Contributions of the American Entomological Institute

L

NVINOSHLINS S31uvVyuaIT LIB

1]

INSTITUTION
S3I¥VYUSI)

S3luvug

NYINOSHLIWS
MITHSONIAN
NVINOSH.LINS

NVINOSHLIWS sa1uvualy LIBRARIES SMITHSONIAN INSTITUTION

NOILNLILSNI

NOILNLILSNI

NOILNLILSNI

INSTITUTION
INSTITUTION

NVINOSHLINS SaluvuaI

AN

NVINOSHLINS S3INVYSIT LIBRARIES S$

SMITHSONIAN
SMITHSONI

INSTITUTION NOILNLILSNI

NOILLALILSNI

Saluvugi LIBRARIES

SAIYVYGIT LIBRARIES
SSIYVYUGIT LIBRARIES

INSTITUTION

INSTITUTION

N

NOILNLILSNI

NVINOSHLIINS
NVINOSHLIWS

NVINOSHLIWS

IBRARIES SMITHSONIA

NOILALILSNI
NOILNLILSNI

NOILNLILSNI

INSTITUTION
saiuvygi

INSTITUTION

= 3 =
- w a
bt x =
oan bs sat |
oa po 2
: es D
w.
mM
< Ww =

NVINOSHLINS Saluvugiy

SMITHSONIAN
NYINOSHLIWS
SMITHSONIAN

INSTITUTION

SMITHSONIAN

¢ 3 x
\ te, = ox
ee ae <<
rena a oc
Be | 2 a
NVINOSHLINS S3tyuvygit
re aBP z .
= Ru ~
> ue >
Sen ee
[Fe o
pr MITHSONIAN INSTITUTION
= ~e “aA. Ss
= WY.
fe = W's. Ry =a
: 2M 2
<= = =
> 7
= = . &
ie poaldva ait
@ m z
a rs =a
ce pay re
i. Si
e) pe oO
=z | as *
_, SMITHSONIAN_INSTITUTION
a +,
a “oy
“ee 6 aie
Zz i |
NVINOSHLINS S3tYVUaIT
wy ne z
= YY; =
+ g fp, 5
6p) vw, WA
oO y Seys Fi Hm
2 ; be
| a =
; 2 | fa?
SMITHSONIAN INSTITUTION NOILN

SAiuvustT LIBRARIES

Ww eee
z
~ &
mc Lo
i a
om.” it.
pie ©
wey 2
i =
wo 2)
| eee
z :
pm
pe 8 | —
os a
Zz

INSTITUTION NOILNLILS

LIBRARI

Saacatt
\

Fk 2th i
+ ce
Sa ay
Det hae
Na

Contributions
of the

American Entomological Institute

Volume 4, Number 1, 1968

“"v"

MOSQUITO STUDIES (Diptera, Culicidae)
X. Dixinae originally described from North America.
By

By T. Michael Peters

XI. Mosquitoes originally described from Argentina,
Bolivia, Chile, Paraguay, Peru, and Uruguay.

By

John N- Belkin, Robert X. Schick, and Sandra J. Heinemann

CONTRIBUTIONS
of the
AMERICAN ENTOMOLOGICAL INSTITUTE

The Contributions are for larger papers on insects. Each paper is a separate number, with

separate pagination and index. Separate numbers aggregating about 500 pages constitute a volume.
Issues appear irregularly, as suitable manuscripts are available.

Copies are sold separately or in subscriptions to complete volumes. Complete volumes are
$12.00. The price of separate numbers varies. Subscribers are billed for each volume with its
beginning number, and receive the parts as issued. Orders for separate numbers that total less
than $8.00 must be accompanied by payment. Address orders or correspondence to the American
Entomological Institute, 5950 Warren Road, Ann Arbor, Michigan 48105, U. S. A.

No.
No.

No.

bide

2a.

3.

2 44

Parts of Volume 1, with prices

Dasch, Clement E., 1964. The Neotropic Diplazontinae (Hymenoptera, Ichneumonidae).
75 pages, 69 figures. Price: $2.25, postpaid
Mosquito Studies (Diptera, Culicidae), 1965.
I. Belkin, John N., etal. A project for a systematic study of the mosquitoes of
Middle America.
II. Belkin, John N., et al. Methods for the collection, rearing and preservation of
mosquitoes.
78 pages, 4figures. Price: $2.25, postpaid
Same as no. 2, but in Spanish. Price: $2.25, postpaid
Matthews, Robert W., 1965. The biology of Heriades carinata Cresson. (Hymenoptera,
Megachilidae) 33 pages, 23 figures. Price: $1.00, postpaid
Mosquito studies (Diptera, Culicidae), 1965.
Ill. Ramalingam, Shivaji, and John N. Belkin. Two new Aedes from Tonga and Samoa.
IV. Belkin, John N. The mosquitoes of the Robinson-Peabody Museum of Salem expedi-
tion to the southwest Pacific, 1956.
34 pages, 3figures. Price: $1.00, postpaid
Mosquito studies (Diptera, Culicidae), 1965.
V. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.
Mosquitoes originally described from Middle America.
95 pages. Price: $2.75, postpaid
Mosquito studies (Diptera, Culicidae), 1966.
VI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.
Mosquitoes originally described from North America.
39 pages. Price: $1.00, postpaid
DeLong, Dwight M. and Paul H. Freytag, 1967. Studies of the world Gyponinae
(Homopotera, Cicadellidae). A synopsis of the genus Ponana.
86 pages, 257 figures. Price: $2.50, postpaid
Contributions to the mosquito fauna of southeast Asia.
I. Delfinado, Mercedes D. The genus Aedes, subgenus Neomacleaya Theobald in
Thailand.
55 pages, 20 figures. Price: $1.75, postpaid

Parts of Volume 2, with prices

Contributions to the mosquito fauna of southeast Asia, 1967.
Il. Bram, Ralph A. The genus Culex in Thailand (Diptera: Culicidae).
296 pages, 104 figures. Price: $9.00, postpaid
Schuh, Toby, 1967. The shore bugs (Hemiptera: Saldidae) of the Great Lakes region.
35 pages, 30 figures. Price: $1.25, postpaid
Snider, Richard J., 1967. The chaetotaxy of North American Lepidocyrtus s. str.
(Collembola: Entomobryidae). 28 pages, 57 figures. Price: $1.00, postpaid —
Contributions to the mosquito fauna of southeast Asia. :
Ill. Delfinado, Mercedes D., 1968. The genus Aedes, subgenus Neomacleaya Theobald
of southeast Asia. 76 pages, 26figures. Price: $2.25, postpaid
Contributions to the mosquito fauna of southeast Asia.
IV. Knight, Kenneth L., 1968. Species of the subgroup Chrysolineatus of group D,
genus Aedes, subgenus Finlaya Theobald. 45 pages, 12 figures. Price: $1.50,
postpaid

Parts of Volume 3, with prices

Mosquito studies (Diptera, Culicidae), 1968.
VII. Belkin, John N. The Culicidae of New Zealand.
182 pages, 30 figures. Price: $5.50, postpaid
Mosquito studies (Diptera, Culicidae), 1968.
VIll. Zavortink, Thomas J. A prodrome of the genus Orthopodomyia.
221 pages, 41 figures. Price: $6.75, postpaid
Fleming, Richard C., 1968. Head musculature of sphinx moths (Lepidoptera:
Sphingidae). 32 pages, 37 figures. Price: $1.00, postpaid
Mosquito studies (Diptera, Culicidae), 1968.
IX. Belkin, John N. The type specimens of New World mosquitoes in European
museums. 69 pages. Price: $2.00, postpaid

Parts of Volume 4, with prices

Mosquito studies (Diptera, Culicidae), 1968.
X. Peters, T. Michael. Dixinae originally described from North America.
7 pages.
XI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann. Mosquitoes
originally described from Argentina, Bolivia, Chile, Paraguay, Peru, and Uruguay.
29 pages. Price of X and XI: $1.25, postpaid

MOSQUITO STUDIES (Diptera, Culicidae)
X. DIXINAE ORIGINALLY DESCRIBED FROM
NORTH AMERICA!

By

T. Michael Peters“

Belkin, Schick and Heinemann (1966) presented all the available data on
the source of the type material for all the nominal species of Culicinae and
Chaoborinae originally described from North America north of Mexico. In
the present paper the treatment of the family Culicidae in the broad Sense is
completed for North America by similar information for the Dixinae. The data
are presented in the same manner. Included in the list are species described
by Peters in Peters and Cook (1966), whose taxonomic treatment is followed
here.

LIST OF SPECIES

Dixinae

The habitats of the immature stages of the Dixinae have been recorded for
only 16 species in North America. This information is given below under BIO-
NOMICS for each of these species. For all the other species the following in-
formation is presented here as a guide to locating breeding sites. In general
the larvae and pupae are found among vegetation or debris along the margins of
various bodies of ground water. They are particularly abundant on the margins
of very small streams, springs or seepages but occur also in pools in larger,
swifter streams aS well as in permanent, semipermanent and temporary stag-
nant pools, ponds, lakes and even in bogs and Swamps. Occasionally they may
be present in artificial containers.

256. Dixa arge Dyar & Shannon, 1924b:199-200. TYPE: Holotype o,
Longmire Springs (Mt. Rainier National Park), Washington, 14 June 1917,
H.G. Dyar (USNM, 27454). BIONOMICS: [Larvae probably among emergent
vegetation in clear, eurythermous, moderately fast, acidic to neutral, shallow
streams or small rivers. |

Pecntibation from project ''Mosquitoes of Middle America" supported by
Public Health Service Research Grant AI-04379 and U.S. Army Medical Re-
search and Development Command Research Contract DA-49-193-MD-2478.

Ee et of Entomology, University of Massachusetts, Amherst, Mass. |
01002.

2 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

257. Dixa blax Dyar & Shannon, 1924b:199. TYPE: Holotype “, Bright
Angel (Coconino Co.), Arizona, 10 May 1903, H.S. Barber (USNM, 27453).

258. Dixa brevis Garrett, 1924a:6. TYPE: Holotype “, McLean Creek,
about 2 mi from mouth, Okanagan Falls, British Columbia, 23 Sept, C. Garrett
(CNC). BIONOMICS: Males occur in swarms flying low along the banks about
a foot off the water. [Larvae probably in shallow acidic seepage pools or hoof-
prints. |

259. Dixa distincta Garrett, 1925a:12. TYPE: Holotype & (130), St. Joseph
Creek, Cranbrook, British Columbia, 16 Oct, C. Garrett (CNC).

260. Dixa fluvica Peters, 1966:242-243. TYPE: Holotype “, Valley Creek
(Washington Co.), Minnesota, 3 July 1962, T.M. Peters (USNM, 67268). BIO-
NOMICS: Adults swept from vegetation near stream. Larvae among duckweed
and watercress in quiet backwater just below entrance of spring into clear, fast,
rocky stream.

261. Dixa fraterna Garrett, 1924a:6. TYPE: Holotype “, Bull River,
British Columbia, 10 Oct 1923, C. Garrett (CNC).

262. Dixa fusca Loew, 1863a:4. TYPE: Lectotype 2, New York, probab-
ly collected by Osten Sacken (MCZ 10391, by PRESENT SELECTION; MCZ
10391 [2] here designated as paratebtatype).

263. Dixa mystica Dyar & Shannon, 1924b:197-198 [=fusca]. TYPE: Lec-
totype &, Cabin John (Montgomery Co. ), Maryland, 22 Feb 1915, R.C. Shannon
(USNM, 27451, by PRESENT SELECTION; & adult, Difficult Run (Fairfax Co.),
Virginia, 3 Oct 1915, W.R. McAtee, USNM, 27451 is here designated paralec-
totype).

264. Dixa hegemonica Dyar & Shannon, 1924b:194-195. TYPE: Holotype
co, Eureka (Humboldt Co.), California, 25 May 1903, H.S. Barber (USNM,
27446).

265. Dixa inextricata Dyar & Shannon, 1924b:198-199. TYPE: Lectotype
o, (slide 2008), Dead Run (Fairfax Co.), Virginia, 25 Mar 1914, R.C. Shannon
(USNM, 27452; selection of Peters and Cook, 1966:245).

266. Dixa johannseni Garrett, 1924a:7. TYPE: Holotype “, Michel, Brit-
ish Columbia, 21 Sept, C. Garrett (CNC). BIONOMICS: [Adults probably swarm
above streams. |

267. Dixa lobata Garrett, 1924a:6. TYPE: Holotype “, Cranbrook, Brit-
ish Columbia, 30 June, C. Garrett (CNC).

268. Dixa melanderi Peters, 1966:246. TYPE: Holotype , Puget, Wash-
ington, 4 July 1924, A.L. Melander (USNM, 67269).

269. Dixa modesta Johannsen, 1903b:429-431. TYPE: Lectotype <, Ithaca
(Tompkins Co.), New York, Apr 1902, O.A. Johannsen (Cornell 2992, by
PRESENT SELECTION). BIONOMICS: Larvae in slow flowing streams.

270. Dixa neohegemonica Peters, 1966:247. TYPE: Holotype “, Mill
Valley (Marin Co.), California, 6 May 1936 (CAS).

271. Dixa notata Loew, 1863a:4. TYPE: Lectotype 2, Maryland, probab-
ly collected by Osten Sacken (MCZ 10393, by PRESENT SELECTION; MCZ
10393 [2] is here designated paralectotype).

272. Dixa pullogruma Peters, 1966:248-249. TYPE: Holotype 0, Rapid
Creek, 2 mi E of Rochford (Pennington Co.), South Dakota, 14 Sept 1961, T.M.
Peters et al. (USNM, 67270). BIONOMICS: Larvae ina fast, cold stream.

273. Dixa repanda Peters, 1966:249. TYPE: Holotype “, Goodhue Co.,
Minnesota, 21 May 1937, D.G. Denning (USNM, 67271).

274. Dixa rhathyme Dyar & Shannon, 1924b:197. TYPE: Holotype ~,
Glacier (Whatcom Co.), Washington, 4 June 1917, H.B. Dyar (USNM, 27450).

Peters: Topotypic North American Dixinae a

275. Dixa montana Garrett, 1924a:5; garretti Cooper & Rapp, 1944:247,
new name [=rhathyme|. TYPE: Holotype , Cranbrook, British Columbia, 30
June, C. Garrett (CNC).

276. Dixa parva Garrett, 1924a:6. [=rhathyme]. TYPE: Bee 7.
Wasa, British Columbia, Oct, C. Garrett (CNC).

277. Dixa rudis Garrett, 1924a:6. TYPE: Holotype ¢, Ta Ta eee?
Wasa, British Columbia, 27 Sept, C. Garrett (CNC).

278. Dixa similis Johannsen, 1923a:57-58. TYPE: Holotype “, Ithaca
(Tompkins Co.), New York, Aug, O.H. Johannsen (Cornell 2993).

279. Dixa terna Loew, 1863a:3. TYPE: Lectotype o, New York, probab-
ly collected by Osten Sacken (MCZ 10394, by PRESENT SELECTION: MCZ
10394 [2] is here designated as paralectotype).

280. Dixa venosa Loew, 1872a:50. TYPE: Holotype 2, Belfrage, Texas
-(MCZ 10395).

281. Dixa xavia Dyar & Shannon, 1924b:195-196. TYPE: Holotype “, Los
Gatos (Santa Clara Co.), California, 16 Feb 1906, J.M. Aldrich (USNM, 27448).

282. Dixella aliciae (Johannsen, 1924b:45-46). TYPE: Holotype &, Clare-
mont, across street E from High School (Los Angeles Co.), California, 1 May
1923, Alica M. Westfall (Cornell 2995). BIONOMICS: Reared from larvae
collected from surface of pool formed by leak in irrigation pipe.

283. Dixella aurora Peters, 1966:254-255. TYPE: Holotype “, swamp,
Ringwood Reservoir (Tompkins Co.), New York, 16-17 June 1963, W.W. Wirth
(USNM, 67272).

284. Dixella californica (Johannsen, 1923a:54-55). TYPE: Holotype &,
Stanford University (Santa Clara Co.), California, Mar (Cornell 2770). BIO-
NOMICS: [Larvae probably in clear, shallow, eurythermous, acidic pools with
muddy bottoms and emergent vegetation. |

285. Dixella thones (Dyar & Shannon, 1924b:193-194) [=californica].
TYPE: Lectotype oc’, Longmire Springs [Mt. Rainier Nat. Park], Washington,
2 Aug 1905, J.M. Aldrich (USNM, 27445, slide 2028, by PRESENT SELEC-
TION; USNM, 27445, 14 June 1917, H.G. Dyar, is here designated paralecto-
type).

286. Dixella clavata (Loew, 1869b:2). TYPE: Lectotype 2, Lenox (Berk-
shire Co.), Massachusetts, Osten Sacken (MCZ 10390, by PRESENT SELEC-
TION; MCZ 10390 [2] is here designated a paralectotype. )

287. Dixella cornuta (Johannsen, 1923a:55-56). TYPE: Holotype “, Mc-
Lean (Tompkins Co.), New York, Oct (Cornell 2990). BIONOMICS: [Larvae
probably in permanent open or temporary woodland ponds, acidic to neutral,
less than 6 feet deep, 11-28°C, with muddy or leafy bottom. May also occur
in slow to fast, shallow, slightly acidic, 14-22°C stream. |

288. Dixella plexipus (Garrett, 1925b:5) [=cornuta]. TYPE: Holotype &
(150), Nelson, British Columbia, June, C. Garrett (CNC).

289. Dixella simplex (Garrett, 1925a:12) [=cornuta]. TYPE: Lectotype
3, Cranbrook, British Columbia, June, C. Garrett (CNC; by PRESENT SE-
LECTION).

290. Dixella somnolenta (Dyar & Shannon, 1924b:195) [=cornuta]. TYPE:
Holotype &, Viola (Latah Co.), Utah, 21 Aug 1912, J.M. Aldrich (USNM,
27447).

291. Dixella spiralis (Garrett, 1924a:1-2) [=cornuta]. TYPE: Lectotype
o, Wasa, British Columbia, 22 Oct 1923, C. Garrett (CNC 5332, by PRESENT
SELECTION; o, Wasa, British Columbia, 25 Oct 1923 (CNC 5341); o&, Wasa,
British Columbia, 21 Oct 1923; o', Wasa, British Columbia, 23 Oct 1923;

4 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

o, Cranbrook, British Columbia, 2 July 1920 (CNC 2618); 0, Cranbrook, Brit-
ish Columbia, 21 Oct 1923 (CNC 5511) are here designated paralectotypes).

292. Dixella deltoura Peters, 1966:257-258. TYPE: Holotype 0, Nenana,
Alaska, 13 June 1951, R.I. Sailor (USNM, 67274).

293. Dixella dorsalis (Garrett, 1924a:4). TYPE: Holotype &, Wasa, Brit-
ish Columbia, 23 Oct, C. Garrett (CNC). BIONOMICS: [Larvae probably in
shallow bogs, seepage pools, temporary woodland pools, or permanent open
ponds with a temperature range of 13-239C, muddy or leafy bottoms, and 5.0-
6.2 pH. Larvae probably situated at bases of sphagnum or sedge hummocks.
Larvae may also occur in large, slow streams less than 4 ft. deep, with a pH
range of 6.2-7.0, clear, and with a muddy or sandy rock bottom. |

294. Dixella dyari (Garrett, 1924a:3). TYPE: Holotype 0, Cranbrook,
British Columbia, 25 June, C. Garrett (CNC).

295. Dixella indiana (Dyar, 1925:217-218). TYPE: Holotype o, Fort Ben-
jamin Harrison (Marion Co.), Indiana, 23 June 1925, H.K.B. Hufford (USNM,
28419). BIONOMICS: [Larvae probably in clear to slightly turbid seepage pools
in Slow to moderately fast, clear to rust-colored streams with a range of tem-
perature of 15.5-23.5°9C, pH 5.5-8.0, depth of 10 inches-5 feet, muddy to
rocky bottom. Adults probably swarm at edges of stream at twilight. |

*296. Dixella marginata (Loew, 1863a:2). TYPE: Holotype 9, District of
Columbia, Osten Sacken (MCZ 10392).

297. Dixella neoaliciae (Garrett, 1924a:2-3). TYPE: Holotype “, Boze-
man (Gallatin Co.), Montana, 3 Sept, C. Garrett (CNC). BIONOMICS: [Larvae
probably in eddies of cold, clear, slightly acidic streams or in warm, clear,
Slightly acidic ponds less than 3 feet deep and with silted bottom. Adults prob-
ably rest in emergent vegetation during the day. |

298. Dixella nova (Walker, 1848:85). TYPE: o, New York Factory, [Hud-
son's Bay at mouth of Hayes River, Manitoba] (BM). BIONOMICS: [Larvae
probably in shallow, rust-colored bogs and clear temporary woodland pools or
seepage pools with leafy bottoms, a temperature range of 13-17. 5°C and pH of
6.0-6. 8. |

299. Dixella centralis (Loew, 1863a:3) [=nova]. TYPE: Holotype 0, New
York, Osten Sacken (MCZ 10389). Terminalia of the type is missing.

300. Dixella nocheles (Dyar & Shannon, 1924b:196-197) [=nova]. TYPE:
Holotype o, Biscayne Bay (Dade Co.), Florida, probably A. T. Slossen (USNM,
27449).

301. Dixella occidentalis (Garrett, 1924a:4-5) [=nova]. TYPE: Holotype
o, Gold Creek, Cranbrook, British Columbia, 1 May, C. Garrett (CNC).

302. Dixella serrata (Garrett, 1924a:2). TYPE: Holotype o, Oliver, Brit-
ish Columbia, 6 May, C. Garrett (CNC). BIONOMICS: [Larvae probably in
shallow ponds or very slow-moving streams with silt bottoms, a temperature
range of 14.5-28°C and pH of 6.8-7.0. Infrequently larvae may be found in
clear, fast, warm, rocky, very acidic streams. |

303. Dixella techana Peters, 1966:263. TYPE: Holotype “, Hassayampa
River, Wickenburg (Maricopa Co.), Arizona, 29 June 1953, W.W. Wirth
(USNM, 67275). BIONOMICS: [Larvae probably in clear, warm, alkaline,
shallow stream with current less than 50 ft/min. and a rocky bottom. |

304. Dixella universitatis (Cockerell, 1926:166). TYPE: Holotype §,
Boulder (Boulder Co.), Colorado, 27 Sept 1926, Scott Gale (USNM, 29540).

305. Dixella vespertina Peters, 1966:264. TYPE: Holotype “, McFadden
Pond, Corvallis (Benton Co.), Oregon, 17 Sept 1953, V. Roth (Oregon State
Univ. ).

Peters: Topotypic North American Dixinae 3)

*306. Meringodixa chalonensis Nowell, 1951:299. TYPE: Lectotype <,
Chalone Creek [Pinnacles Nat. Monument], California, 20 Apr 1950, W.R.
Nowell (USNM, by PRESENT SELECTION). BIONOMICS: Larvae in protected,
shaded spots among rocks in stream.

Nomen Dubium

307. Dixa recens Walker, 1848:85. TYPE: ¢, New York Factory [Hud-
son's Bay at mouth of Hayes River, Manitoba]; type lost and the description is
inadequate for determination of the taxonomic species involved (Peters and
Cook, 1966:251).

LIST OF LOCALITIES

Canada

BRITISH COLUMBIA

Bull River: 261. Dixa fraterna.

Cranbrook: 259. Dixa distincta; 267. Dixa lobata; 294. Dixella dyari; 275.
Dixella garretti; 301. Dixella occidentalis; 289. Dixella simplex; 291. Dixella

Spiralis.
Michel: 266. Dixa johannseni.

Nelson: 288. Dixella plexipus.

Okanagan Falls: 258. Dixa brevis.

Oliver: 302. Dixella serrata.

Wasa: 276. Dixa parva; 277. Dixa rudis (Ta Ta Creek); 293. Dixella dor-
salis.

MANITOBA
New York Factory (Hudson's Bay at mouth of Hayes River): 298. Dixella
nova; 307. Dixa recens.

United States

ALASKA
Nenana: 292. Dixella deltoura.

ARIZONA
Bright Angel (Coconino Co.): 257. Dixa blax.
Wickenburg, Hassayampa River (Maricopa Co.): 303. Dixella techana.

CALIFORNIA
Chalone Creek (San Benito Co.): 306. Meringodixa chalonensis.
Claremont (Los Angeles Co.): 282. Dixella aliciae.
Eureka (Humboldt Co.): 264. Dixa hegemonica.
Los Gatos (Santa Clara Co.): 281. Dixa xavia.
Mill Valley (Marin Co.): 270. Dixa neohegemonica.
Stanford University (Santa Clara Co.): 284. Dixella californica.

6 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

COLORADO
Boulder (Boulder Co.): 304. Dixella universitatis.

DISTRICT OF COLUMBIA
Washington: 296. Dixella marginata.

FLORIDA
Biscayne Bay (Dade Co.): 300. Dixella nocheles.

IDAHO |
Viola (Latah Co.): 290. Dixella somnolenta.

INDIANA
Fort Benjamin Harrison (Marion Co.): 295. Dixella indiana.

MARYLAND
Cabin John (Montgomery Co.): 263. Dixa mystica.
Locality not specified: 271. Dixa notata.

MASSACHUSETTS
Lenox (Berkshire Co.): 286. Dixella clavata.

MINNESOTA
Valley Creek (Washington Co.): 260. Dixa fluvica.
Goodhue Co.: 273. Dixa repanda.

MONTANA
Bozeman (Gallatin Co.): 297. Dixella neoaliciae.

NEW YORK
Ithaca (Tompkins Co.): 269. Dixa modesta; 278. Dixa similis.
McLean (Tompkins Co.): 287. Dixella cornuta.
Ringwood Reservoir (Tompkiys Co.): 283. Dixella aurora.
Locality not specified: 262. Dixa fusca; 279. Dixa terna; 299. Dixella
centralis.

OREGON
Corvallis, McFadden Pond (Benton Co.): 305. Dixella vespertina.

SOUTH DAKOTA
Rapid Creek (Pennington Co.): 272. Dixa pullogruma.

TEXAS
Belfrage: 280. Dixa venosa.

VIRGINIA
Dead Run (Fairfax Co.): 265. Dixa inextricata.

WASHINGTON

Glacier (Whatcom Co.): 274. Dixa rhathyme.

Mt. Rainier National Park, Longmire Springs: 256. Dixa arge; 285. Dix-
ella thones.

Puget: 268. Dixa melanderi.

Peters: Topotypic North American Dixinae 7

REFERENCES CITED

The citations of author, date and page in the List of Species refer to the
bibliography in 'A Catalog of the Diptera of America North of Mexico" (U.S.
D.A. Agricultural Handbook No. 276, 1965).

Belkin, J.N., R.X. Schick and S.J. Heinemann
1966. Mosquito Studies (Diptera, Culicidae). VI. Mosquitoes originally de-
scribed from North America. Amer. Entomol. Inst., Contrib. 1(6).
oo Dp.
Peters, T.M.
1966. [Descriptions of new species] In Peters, T.M. and E. F. Cook, 1966
(see below).
Peters, T.M. and E. F. Cook
1966. The Nearctic Dixidae (Diptera).
lications 5(5):231-278. 23 Dec.

Entomol. Soc. Amer., Misc. Pub-

INDEX TO SPECIFIC NAMES

The references are to the numbers preceding the species names in the

Rist of Species.

aliciae, 282
arge, 256
aurora, 283
blax, 257
brevis, 258
californica, 284
centralis, 299
chalonensis, 306
clavata, 286
cornuta, 287
deltoura, 292
distincta, 259
dorsalis, 293
dyari, 294
fluvica, 260
fraterna, 261
fusca, 262
garretti, 275
hegemonica, 264
indiana, 295
inextricata, 265
johannseni, 266
lobata, 267
marginata, 296
melanderi, 268
modesta, 269

montana, 275
mystica, 263
neoaliciae, 297
neohegemonica, 270
nocheles, 300
notata, 271
nova, 298
occidentalis, 301
parva, 276
plexipus, 288
pullogruma, 272
repanda, 273
recens, 307
rhathyme, 274
rudis, 277
serrata, 302
similis, 278
simplex, 289
somnolenta, 290
spiralis, 291
techana, 303
terna, 279
thones, 285
universitatis, 304
venosa, 280
vespertina, 305
xavia, 281

£ eeatine

—

yet
Pyar stti

MOSQUITO STUDIES (Diptera, Culicidae)

XI. MOSQUITOES ORIGINALLY DESCRIBED FROM ARGENTINA,

BOLIVIA, CHILE, PARAGUAY, PERU, AND URUGUAY!
By

John N. Belkin, Robert X. Schick and Sandra J. Heinemann?

This is the fourth in a series of papers providing information on the source
of the original type material of mosquitoes described from the Americas (Bel-
kin, Schick and Heinemann, 1965,1966; Peters, 1968). The concluding paper
on species described from Brazil and unspecified localities in South America is
in preparation. For explanation of the arrangement and method of presentation,
the first paper of the series should be consulted.

We are most grateful to Osvaldo H. Casal of the Instituto Nacional de Mi-
crobiologfa, Buenos Aires, for his careful study of the type material in Argen-
tina and for allowing us to publish his lectotype designations in this paper. We
also thank Alan Stone for information on material contained in the U.S. Nation-
al Museum.

ARGENTINA

List of Species

1. Anopheles (A.) annulipalpis Lynch Arribalzaga, 1878. TYPE: ¢, Bara-
dero (Buenos Aires), Apr 1878 (NE, according to O.H. Casal). BIONOMICS:
[Larvae in semipermanent turbid ground water with scanty vegetation. |

2. Anopheles (A.) holmbergi Del Ponte & Heredia, 1945 [=fluminensis].
TYPE: Holotype & (378) with genitalia slide (2293), Corpus, San Ignacio (Mis-
iones), May 1944, R. L. Heredia (INM). BIONOMICS: [Larvae probably in per-
manent or semipermanent ground waters. |

*3. Anopheles (A.) argentinus (Bréthes, 1912) [=pseudopunctipennis]. TYPE:
Lectotype 0, bearing label by Brethes, /Proterorhynchus argentinus Brethes,
type, Tucumadn/ (BA; PRESENT DESIGNATION by O.H. Casal). BIONOMICS:
[Larvae in small streams in mountains and foothills; water fresh, with mode-
rate flow, and containing algae. |

1 Contribution from project ''Mosquitoes of Middle America’' supported by
Public Health Service Research Grant AI-04379 and U.S. Army Medical Re-
search and Development Command Research Contract DA-49-193-MD-2478.

“Department of Zoology, University of California, Los Angeles, Califor-
nia 90024.

10 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

4. Anopheles (A.) tucumanus Lahille, 1912 [=pseudopunctipennis|. TYPE:
Holotype 2, Rio Sali (Tucuman), Dinelli (NE, according to O.H. Casal). BIO-

NOMICS: [As for 3. argentinus. |

5. Anopheles (A.) patersoni Alvarado & Heredia, 1947 [=ssp. of pseudo-
punctipennis|. TYPE: Eggs, Tucuman (probably NE, according to O. H. Casal).
BIONOMICS: [As for 3. argentinus. |

*6. Anopheles (Nyssorhynchus) albitarsis Lynch Arribalzaga, 1878. TYPE:
?, Baradero (Buenos Aires) [NE; original type material lost; redescription by
Umana, Heredia and Siquot (1960:609-618) from type locality does not constitute
a neotype designation, this material also lost]. BIONOMICS: [Larvae usually
in mats of aquatic vegetation in large ponds, marshes and lagoons of overflow-
ing rivers, and in large sunlit ground pools with vegetation. |

7. Anopheles (N.) rooti (Brethes, 1926) [=argyritarsis]. TYPE: o, Tu-
cuman, N.C. Davis; only remaining material genitalia slide in poor condition,
without sidepiece, with Bréthes label, ''Cellia Rooti Bréthes=argyritarsis auct.
(nec R.D.), Tucuman, N.C. Davis dedit, III. 1956'"' (BA). BIONOMICS: [Larvae
in all types of sunlit and partially shaded ground waters.|

8. Anopheles (N.) evansae (Bréthes, 1926). TYPE: Lectotype 0 genitalia
slide found in INM (returned to BA), with following Brethes label, ''Cellia ev-
ansi Brethes=tarsimaculata auct. (nec Goeld), Tucuman, N.C. Davis ded., III.
1926, Pr. micr. An.M.," in Del Ponte's hand '384" "DP14,'"" 'DP14= ? noroes-
tensis G. y L., 1938, [signature in ink], Tipo de evansi [signature in black pen-
cil] (BA; PRESENT DESIGNATION of O.H. Casal). BIONOMICS: [Larvae in
marshes and marshy margins of clear freshwater pools, streams and lakes.|

9. Anopheles (N.) clarki Komp, 1943 [=noroestensis]|. TYPE: Holotype
o& with genitalia slide, Monteros (Tucuman), June 1940, C.A. Alvarado (USNM,
56476; apparently lost, not mentioned in Stone and Knight, 1956b). BIONOM-
ICS: [Larvae probably in shaded fresh water in wooded swamps, pools or stag-
nant streams.|

*10. Anopheles (N.) bachmanni Petrocchi, 1925 [=ssp. of triannulatus].
TYPE: o, 2, Corrientes, Entre Rios, Formosa (NE, according to O. H. Casal).
BIONOMICS: [Larvae probably in freshwater pools and lakes, and along river
margins, associated with vegetation. |

11. Anopheles (N.) perezi Shannon & Del Ponte, 1928 [=triannulatus bach-
manni]. TYPE: 9, [Finca] Santa Barbara, Departamento de la Capital (Tucu-_
man) (NE, according to O.H. Casal). BIONOMICS: Larvae in a lake.

12. Anopheles (N.) davisi Paterson & Shannon, 1927 [=triannulatus bach-
manni]. TYPE: o, @, larvae, Tres Pozos, [near] Embarcacion (Salta), 19-21
Apr 1927, Paterson, Shannon and Shannon (USNM, see Stone and Knight, 1956b:
277; INM, 1 2 with Shannon's labels, ''House, Embarcacion, Salta, 21.4. 27,
R.C. Shannon/Anopheles (Nyssorhynchus) davisi Pat. & Snn (Cotype)'"'). BIO-
NOMICS: Larvae in very large masses of green algae in a lake.

13. Toxorhynchites (Lynchiella) cavalierii Garcia & Casal. TYPE: Holo-
type 2 with associated larval and pupal skins (c94), Puerto Iguazu about 50 m
from the Hosteria Ruffino (Igu 15) (Misiones), 25 June 1965, Casal, Hepper
and Garcia (INM). BIONOMICS: Larvae in a bromeliad 1 m above the ground.

14. Toxorhynchites (L.) tucumanus (Brethes, 1926) [=guadeloupensis].
TYPE: Lectotype & (1 of 6 with type labels) with following Brethes labels, ''Tu-
cuman, 10-VII-926, E.G. Cabarrou/elevé chez moi/Megarhinus tucumanus
Brethes" (BA; PRESENT DESIGNATION of O.H. Casal). BIONOMICS: [Lar-
vae in leaf axils of bromeliads. |

15. Toxorhynchites (L.) arborealis (Shannon & Del Ponte, 1928) [=guade-

Belkin et al: Topotypic South American Mosquitoes i

loupensis]. TYPE: 9, Lules (Tucuman), Shannon and Del Ponte, 22 Mar 1927
[NE, according to O.H. Casal]. BIONOMICS: Larvae in a treehole.
16. Toxorhynchites (L.) separatus (Lynch Arribalzaga, 1891) [=ssp. of
haemorrhoidalis}|. TYPE: Lectotype “ with Lynch label ''Megarhina separata
n.sp.'’ and Del Ponte label ''Megarhinus haemorroidalis Fabr., D. P., X-49,"
according to original description collected in Formosa by E. L. Holmberg (BA;
PRESENT DESIGNATION by O.H. Casal). BIONOMICS: [Larvae probably in
leaf axils of bromeliads. |

17. Toxorhynchites (L.) lynchi (Dyar & Knab, 1906) [=haemorrhoidalis sep-
aratus|. TYPE: Lectotype 0, only specimen in collection with Lynch's label,
"Megarhyna haemorroidalis Fabr.,'' also with Del Ponte's label, 'Megarrhynus
lynchi tipo, D. P. X-49,"' according to original description collected in Formo-
sa by E.L. Holmberg (BA; PRESENT DESIGNATION by O.H. Casal). BIO-

NOMICS: [As for 16. haemorrhoidalis separatus.|

18. Trichoprosopon (Runchomyia) paranensis (Bréthes, 1910) [Distinct spe-
cies, according to O.H. Casal]. TYPE: Lectotype 9, the better of the 2 speci-

mens bearing the following label, ''Tuyuparé, 25.1.1908, J.B. /Lynchiaria pa-
ranensis" (BA; PRESENT DESIGNATION by O.H. Casal). BIONOMICS: [Lar-
vae probably in leaf axils of cattails, Typha.|

19. Wyeomyia (Nunezia) lateralis Petrocchi, 1927. TYPE: 92, Zapla (Ju-
juy), 16 Mar 1916, A. Neiva (NE, according to O.H. Casal). BIONOMICS: [Lar-
vae probably in leaf axils of bromeliads.|

*20. Wyeomyia (Davismyia) petrocchiae (Shannon & Del Ponte, 1928). TYPE:
Holotype 2, Raco (Tucuman), 13 Feb 1927, Shannon and Del Ponte (USNM; see
Stone and Knight, 1957b:126). BIONOMICS: [Larvae probably in treeholes.|

21. Wyeomyia (Davismyia) monoleua (Martini, 1931)[=petrocchiae]. TYPE:
Holotype ¢, San Jose (Formosa), Oct 1925 (SMNS). BIONOMICS: [Larvae prob-
ably in treeholes.|

22. Wyeomyia (Menolepis) leontiniae (Bréthes, 1910) [=leucostigma]. TYPE:
Lectotype 2, the best of the 5 9, all without type label and all bearing Bréthes
labels, ''Tuyupare, 25-I-1908, J.B" and ''Limatus Leontiniae" (BA; PRESENT
DESIGNATION of O.H. Casal). BIONOMICS: [Larvae probably in leaf axils of
cattails, Typha.|

23. Wyeomyia (Dendromyia) belkini Casal & Garcia, 1966. TYPE: Holo-
type & with genitalia, larval and pupal skins (C57), intersection of the arroyo
Ibicuy and route 101, about 20 km west of Cataratas (Igu 26) (Misiones), 26
June 1965, Hepper, Garcia and Casal (INM). BIONOMICS: Adults taken in the
forest interior during the day, probably attracted to a human host. Larvae in
bamboo internodes.

24. Wyeomyia (D.) typharum (Shannon & Del Ponte, 1928) [=melanocepha-
la]. TYPE: Lectotype 9, bearing label probably in Del Ponte's hand 'Dendro-
myia typharum" and also with following labels (present on other @), "Jujuy,
Ledesma, 3.19.26/Davis and Shannon/49036" and ''Metatypus"” (BA; PRESENT
DESIGNATION of O.H. Casal). BIONOMICS: Larvae in leaf axils of cattails,
Typha.

25. Phoniomyia muehlensi (Petrocchi, 1927). TYPE: Holotype, sex un-
known, Chaco, Oct-Nov 1924, J. Petrocchi (NE, according to O. H. Casal). BI-
ONOMICS: [Larvae in leaf axils of bromeliads.]|

26. Limatus exhibitor Shannon & Del Ponte, 1928 [=durhamii]. TYPE:
Lectotype & with following label, "Iguazu, Mis.[iones] 18.6.27, Shannon & Del
Ponte" and in Shannon's hand, ''Limatus exhibitor Snn & D. P’’ (INM; PRESENT
DESIGNATION of O. H. Casal). BIONOMICS: [Larvae in broken bamboo and in
artificial containers at the type locality (O. H. Casal).|

£2 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968 .

27. Sabethes (S.) neivai Petrocchi, 1927 [=albiprivus]|. TYPE: Holotype 9,
Santa Clara (Jujuy), May 1916, Arturo Neiva (NE, according to O.H. Casal).
BIONOMICS: [Larvae probably in treeholes or bamboo internodes. |

28. Coquillettidia (Rhynchotaenia) fasciolata (Lynch Arribalzaga, 1891).
TYPE: Originally described from 2 from Navarro (Buenos Aires), Mar 1886,
F. Lynch (NE), neotype ?, Parque Lezama, Capital Federal (Buenos Aires), 13
Dec 1928, Carillo (INM; designation of Castro and Bressanello, 1952:232). BI-
ONOMICS: [Larvae probably on rootlets of grasses and herbaceous vegetation
in mud and fine sediment in very shallow water on the margins of swamps,
ponds or streams. |

29. Coquillettidia (Rhynchotaenia) araozi (Shannon & Del Ponte, 1928) [=
venezuelensis]|. TYPE: ?, Ledesma (Jujuy), 20 Mar 1926, Davis and Shannon;
no specimens of type series in BA, according to O. H. Casal (USNM; lectotype
to be designated by Alan Stone). BIONOMICS: [Larvae probably as for 28. fas-
ciolata.| ¢ taken on animal bait.

30. Uranotaenia lanei Martinez & Prosen, 1953. TYPE: Holotype ¢, For-
mosa (BA, according to O.H. Casal). BIONOMICS: [Larvae probably in per-
manent or semipermanent ground waters. |

31. Uranotaenia monilis Shannon & Del Ponte, 1928. TYPE: Lectotype ?
bearing the labels "Resistencia, Chaco, 20.2.27, R.C. Shannon/Uranotaenia
monilis Snn. & D.P.' in Shannon's hand (INM; PRESENT DESIGNATION by
O.H. Casal). BIONOMICS: [Larvae probably in large lowland swamps or in
other sunlit ground waters with some aquatic vegetation. |

*32. Uranotaenia nataliae Lynch Arribalzaga, 1891. TYPE: Lectotype adult,
only the thorax and first abdominal segment remaining, with a label by Lynch
"Uranotaenia nataliae n.sp., Tigre" [at the confluence of the Rio Las Conchas
and Rio Parana (Buenos Aires), Enrique Lynch, brother of Félix] (BA; PRES-
ENT DESIGNATION by O.H. Casal). BIONOMICS: [Larvae probably in large
open swamps with abundant aquatic vegetation such as Pistia and sedges. |

33. Uranotaenia capitis Shannon & Del Ponte, 1928. TYPE: Holotype 2,
Ledesma (Jujuy), 20 Mar 1926, Davis (USNM). BIONOMICS: [As for 32. natal-
iae.|

*34, Uranotaenia pulcherrima Lynch Arribalzaga, 1891. TYPE: , [prob-
ably] bank of Rio Lujan (Buenos Aires), E. L. Holmberg (NE, according to O. H.
Casal). BIONOMICS: [Larvae probably in swamps, stream margins or other
permanent ground waters. | : |

35. Uranotaenia urania Shannon & Del Ponte, 1928 [=pulcherrima]. TYPE:
Lectotype o (2355) with genitalia slide, Resistencia (Chaco), 20 Feb 1927, Shan-
non and Del Ponte (USNM, designation of Stone and Knight, 1957c:200). BIONOM-
ICS: [As for 34. pulcherrima.|

36. Uranotaenia elnora Paterson & Shannon, 1927 [ssp. of pulcherrima].
TYPE: Holotype 2, Tres Pozos near Embarcacion (Salta), 19 Apr 1927, Elnora
S. Shannon (USNM). BIONOMICS: [Probably as for 34. pulcherrima. |

*37. Aedeomyia squamipennis (Lynch Arribalzaga, 1878). TYPE: o, ¢, Ba-
radero (Buenos Aires), Apr 1878, F. Lynch (NE, according to O.H. Casal).
BIONOMICS: [Larvae in lakes, ponds and permanent ground pools with abundant
vegetation and algae. |

38. Psorophora (P.) lynchi Brethes, 1916 [=ciliata]. TYPE: Lectotype 9,
the best of the 4 female syntypes in the BA, and in good condition, bearing the
labels ''B.[uenos] Aires, 20.11.1916, J.B. /Psorophora Lynchi Br." (the other
3 females are similarly labeled) (BA; PRESENT DESIGNATION by O.H. Casal).
BIONOMICS: [Larvae probably in unshaded temporary rain pools. |

Belkin et al: Topotypic South American Mosquitoes 13

‘39. Psorophora (P.) holmbergii Lynch Arribalzaga, 1891. TYPE: Lecto-
type 2, the one specimen in the collection of the BA bearing the following label
by F. Lynch, 'Psorophora Holmbergi n.sp., Chaco'' [Formosa], E.L. Holm-
berg (BA; PRESENT DESIGNATION by O.H. Casal). BIONOMICS: [Larvae in
temporary rain pools.|

40. Psorophora (P.) agoggylia Dyar, 1922 [=holmbergii]. TYPE: Lecto-
type 2, Rio Tapenaga, Colonia Florencia, Gran Chaco (Formosa [Santa Fe)]),
1903, E.R. Wagner (USNM, 25756; designation of Stone and Knight, 1955:286).
BIONOMICS: [Larvae probably in temporary rain pools.|

41. Psorophora (P.) stigmatephora Dyar, 1922 [=pallescens]. TYPE: Lec-
totype & (1660) with genitalia slide, Rio Tapenga, Colonia Florencia, Gran
Chaco (Formosa [Santa Fe]), 1903, E.R. Wagner (USNM, 25756; designation by
Stone and Knight, 1955:286). BIONOMICS: [Larvae probably in temporary rain
pools. |]

42. Psorophora (Janthinosoma) centrale Brethes, 1910 [=ferox]. TYPE:
Lectotype 2 with the following labels, "Buenos Aires, 4. XII. 1907, B./Janthino-
soma centralis [sic] Bréthes'' in Brethes' hand, type locality cited as "Islas del
Parana, Buenos Aires" in the original description (BA; PRESENT DESIGNA-
TION by O. H. Casal). BIONOMICS: [Larvae probably in temporary rain and
floodwater pools in wooded areas.|

43. Psorophora (J.) chaquensis Paterson & Shannon, 1927 [=lutzii]. TYPE:
Holotype °, Tres Pozos, near Embarcacion (Salta), 20 Apr 1927, probably
Paterson and Shannon (NE, according to Casal). BIONOMICS: [Larvae prob-
ably in temporary rain and floodwater pools in wooded areas.|

44. Psorophora (J.) bruchi Petrocchi, 1927 [=varipes|]. TYPE: 0, 2, Pro-
vincia de Buenos Aires, 1 Apr 1925, C. Bruch (NE, according to O. H. Casal).
BIONOMICS: [Larvae probably in temporary rain and floodwater pools in wooded
areas. |

*45. Psorophora (Grabhamia) confinnis Lynch Arribalzaga, 1891. TYPE:
Lectotype 2, the only female of the original 5 specimens remaining in the BA,
in fairly good condition, with the following label by F. Lynch, ''Taeniorhynchus
confinnis F. Lch., Chaco [Formosa],'' E.L. Holmberg (BA; PRESENT DESIG-
NATION by O.H. Casal). BIONOMICS: [Larvae probably in open temporary
ground pools of all types, especially hoofprints and road ruts.]|

46. Psorophora (G.) paulli (Paterson & Shannon, 1927). TYPE: Lectotype
2, proboscis broken, 2 legs missing and abdomen glued on a card, with the fol-
lowing labels in Shannon's handwriting, ''Tres Pozos, Salta, 20.4.27, Shannon
& Snn/Types Psorophora Paulli Paterson & Shannon" (INM; PRESENT DESIG-
NATION by O.H. Casal). BIONOMICS: [Larvae probably in temporary ground
pools or overflows of ponds or streams. |

47. Aedes (O.) meprai Martinez & Prosen, 1953 [=angustivittatus]|. TYPE:
Holotype 2, Reserva Nacional ''Finca del Rey,’’ Departamento de Anta (Salta),
Nov 1952, Martinez and Prosen (BA). BIONOMICS: [Larvae in temporary rain
pools.] Adults captured on human bait in daytime in forest.

48. Aedes (Ochlerotatus) lynchii (Bréthes, 1910) [=crinifer]. TYPE: Lec-
totype 2, the best of 9 females, in fairly good condition, all from Buenos Aires
with the following labels, "6139/Buenos Aires, J. Brethes 19.1. 1903/Culex
Lynchii Br.,'' the latter two by Brethes (BA; PRESENT DESIGNATION by O. H.
Casal). BIONOMICS: [Larvae probably in temporary rain pools. |

49. Aedes (O.) tapinops (Brethes, 1917) [=crinifer]. TYPE: Lectotype c,
in fairly good condition, with the following labels by Brethes, ''S. Isidro [Buenos
Aires], II.2.1917. J. B./Culex tapinops Bret.'' (BA; PRESENT DESIGNATION

14 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

by O.H. Casal). BIONOMICS: [Larvae probably in temporary rain pools. |

50. Aedes (O.) iguazu Shannon & Del Ponte, 1928 [=crinifer]|. TYPE: o, 9
[Cataratas del Iguazu], Misiones (3 2, without type labels but collected at Igua-
zu Falls by Shannon and Shannon in Oct 1927, are present in the USNM; no
specimens of the type series are in the INM). BIONOMICS: [Larvae probably
in temporary rain pools.|

51. Aedes (O.) araozi Shannon & Del Ponte, 1928 [=milleri]. TYPE: Lec-
totype co’, in fairly good condition, with the following labels in Shannon's hand,
"Quebrada San Lorenzo, Salta, 4. VIII.27, R.C. Shannon/A. 154/Aedes araozi
Shannon & Del Ponte," with genitalia slide (655) (INM; PRESENT DESIGNATION
by O. H. Casal). BIONOMICS: Larvae in a small rockhole covered with dead
leaves very near a small torrential stream.

52. Aedes (O.) patersoni Shannon & Del Ponte, 1928. TYPE: Lectotype §,
in very bad condition, a mesothoracic leg and an antenna are missing and the
mesonotum is largely rubbed, with the following labels, ''San Pedro de Jujuy,
27.4.26/Inst. Bac. Ent. nota 44/Shannon and Shannon" (INM; PRESENT DES-
IGNATION by O.H. Casal; the same specimen improperly designated as Holo-
type by Garcia & Ronderos, 1963:32). BIONOMICS: Larvae in a temporary
swampy area formed by rains.

53. Aedes (O.) raymondi Del Ponte, Castro & Garcia, 1951. TYPE: Holo-
type @ (587), San Pedro (Jujuy), 27 Apr 1926, Shannon and Shannon (INM). BI-
ONOMICS: [Larvae probably in temporary rain pools. |

*54. Aedes (O.) confirmatus (Lynch Arribalzaga, 1891) [=scapularis|. TYPE:
Lectotype @, in good condition, with the following labels by F. Lynch, "Ochle-
rotatus cofirmatus [sic], F Lynch, Navarro [Buenos Aires],' May 1887 (BA;
PRESENT DESIGNATION by O. H. Casal). BIONOMICS: [Larvae in temporary
rain pools.|

55. Haemagogus (Stegoconops) petrocchiae Martinez, Carcavallo & Prosen,
1961 (=ssp. of capricornii]. TYPE: Holotype o, Salvador Mazza (Pocitos),
Departamento General San Martin (Salta), Jan 1960, Martinez and Carcavallo
(BA). BIONOMICS: Females taken biting on human bait at dusk and at night in
woods without artificial lights. [Larvae-probably in treeholes. |

56. Haemagogus (Stegoconops) spegazzinii Brethes, 1912. TYPE: Holotype
? [with the following labels, ''No 727 3/Jujuy/ Haemagogus spegazzinii Brethes"
and 2 slides (3028), one with a wing and the other with genitalia, an antenna and
a palpus], Jujuy, 1907, D. Carlos Spegazzini(BA). BIONOMICS: [Larvae prob-
ably in treeholes.] Adult(s) taken in deep woods.

57. Haemagogus (Stegoconops) uriartei Shannon & Del Ponte [=spegazzinii].
TYPE: Holotype % (128-3) with genitalia slide (2353) and pupal skin slide (V 3),
Vipos (Tucuman), 22 Mar 1927, Shannon and Del Ponte (USNM; see Stone and
Knight, 1955:289). BIONOMICS: Holotype bred from larva in a treehole.

58. Culex (Lutzia) patersoni Shannon & Del Ponte, 1928 [=bigoti]. TYPE:
Lectotype & (2356) with genitalia slide, San Pedro (Jujuy), 27 Apr 1926, Shan-
non and Shannon (USNM; designation of Stone and Knight, 1957a:54). BIONOMICS:
[Larvae in ground pools, rockholes and artificial containers. |

59. Culex (C.) ameliae Casal, 1967. TYPE: Holotype o, with associated
larval and pupal skins (Ga 137-31), San Pedro-El Dorado road, about 10 km
from San Pedro (Misiones), 11 Mar 1966, Garcia and Casal (INM). BIONOM-
ICS: Larvae in an extensive flooded depression in the middle of the forest
crossed by the road.

60. Culex (C.) interfor Dyar, 1928 [=bidens]. TYPE: Lectotype & (532)
with genitalia slide (2364), in train between Tucuman and Jujuy (Jujuy, Salta or

Belkin et al: Topotypic South American Mosquitoes 15

Tucuman), 4 May 1927, M. Kisliuk Jr. (USNM; designation by Stone and Knight
1957a:51). BIONOMICS: [Larvae probably in permanent and semipermanent
ground waters.| Adults taken at light on train.

61. Culex (C.) brethesi Dyar, 1919. TYPE: Holotype “ (on slide with the
genitalia and a female) [San Isidro (Buenos Aires), Apr or May 1916] (BA). BI-
ONOMICS: [Larvae probably in permanent or semipermanent ground waters. |

62. Culex (C.) castroi Casal & Garcia, 1967. TYPE: Holotype & with as-
sociated larval and pupal skins (Ga 289-11), Canal 6 and Parana de las Palmas,
Estacion Agropecuaria Delta del I.N. T. A. (near Otamendi) (Buenos Aires), 27
July 1966, Garcia and Casal (INM). BIONOMICS: Larvae in leaf axils of Eryn-
sium sp.

*63. Culex (C.) dolosus (Lynch Arribalzaga, 1891). TYPE: Lectotype o, in
very bad condition, partially covered by fungi, the head and the posterior ab-
dominal segments missing, with the following label by F. Lynch, ''Heterony-
cha dolosa n.sp.,'' no locality label, the following 5 localities given in the orig-
inal description: Las Conchas, Zarate, Baradero, Navarro and Chacabuco (all
Buenos Aires) (BA; PRESENT DESIGNATION by O.H. Casal). BIONOMICS:
[Larvae in permanent and semipermanent ground waters. |

64. Culex (C.) bonariensis Bréthes, 1916 [=dolosus]. TYPE: Lectotype,
with type label but without locality label, and genitalia slide (V) [San Isidro
(Buenos Aires)|, May (BA; PRESENT DESIGNATION by O.H. Casal). BIO-
NOMICS: [Probably as for 63. dolosus. |

65. Culex (C.) eduardoi Casal & Garcia, 1968. TYPE: Holotype with as-
sociated larval and pupal skins (Ga 314-17), Canal 6 and Parana de las Palmas,
Estacion Agropecuaria Delta de I.N. T.A. (near Otamendi) (Buenos Aires), 10
Aug 1966, Garcia and Casal (INM). BIONOMICS: Larvae in a marshy depres-
sion.

66. Culex (C.) fernandezi Casal, Garcia and Cavalieri, 1966. TYPE: Holo-
type o with associated larval and pupal skins (Ga 85-15), stream in Piquirenda
(Salta), 26 Nov 1965, Fernandez, Hepper, Garcia and Casal (INM). BIONOM-
ICS: Larvae in axils of epiphytic or terrestrial bromeliads.

67. Culex (C.) hepperi Casal & Garcia, 1967. TYPE: Holotype & with as-
sociated larval and pupal skins (Ga 268-10), Canal 6 and Parana de las Palmas,
Estacion Agropecuaria Delta de I.N. T. A. (near Otamendi) (Buenos Aires), 30
Apr 1966, Garcia and Casal (INM). BIONOMICS: Larvae in leaf axils of Eryn-
gium.

68. Culex (C.) lahillei Bachmann & Casal, 1962. TYPE: Holotype & (1124)
with genitalia slide (5219), Achiras (Cordoba), Mar 1938, Del Ponte (INM). BI-
ONOMICS: [Larvae probably in permanent and semipermanent ground waters. |

69. Culex (C.) maxi Dyar, 1928. TYPE: Lectotype o on slide 2360 and
some legs on pin mount, San Pedro (Jujuy), 11 June 1927, M. Kisliuk Jr. (USNM;
designation by Stone and Knight, 1957a:53). BIONOMICS: [Larvae in permanent
and semipermanent ground waters. |

70. Culex (C.) autumnalis Weyenbergh, 1882 [=pipiens quinquefasciatus].
TYPE: o, °, larva, pupa, Rio Primero (Cérdoba) (? Universidad de Cordoba).
BIONOMICS: [Larvae in large artificial containers and contaminated ground
water. |

71. Culex (C.) saltanensis Dyar, 1928. TYPE: Holotype o’, Campo Santo
de Salta (Salta), 12 May 1927, M. Kisliuk Jr. (USNM). BIONOMICS: [Larvae
probably in permanent or semipermanent ground waters. |

*72. Culex (Allimanta) tramazayguesi Duret, 1954. TYPE: Holotypeo" (M.R.
110, E.1), Monte Coman (Mendoza), 16 Dec 1953, Bejarano and Duret (CMPHM).

16 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

BIONOMICS: [Larvae found in hypersaline ground water near a thermal spring
(Casal).]|

73. Culex (Melanoconion) aliciae Duret, 1953. TYPE: Holotype o (Mis. 13,
E 8) with genitalia slide, Cerro Azul, Arroyo San Juan (Misiones), 2 May 1949,
Bejarano and Duret (CMPHM). BIONOMICS: [Larvae probably in stream pools
or other permanent or semipermanent ground waters. |

74. Culex (Mel.) bejaranoi Duret, 1953. TYPE: Holotype o (A. U. 93, E.
20) with genitalia slide, Bernardo de Irigoyen (Misiones), 18 Mar 1951, Duret
(CMPHM). BIONOMICS: [Larvae found in large ground pool in a probably
flooded forest (Casal). |

75. Culex (Mel.) intrincatus Brethes, 1916. TYPE: Holotype <, lowlands
of San Isidro (Buenos Aires), 15 May 1916, Brethes (BA). BIONOMICS: [Lar-
vae probably in permanent or semipermanent ground waters.|

76. Culex (Mel.) martinezi Casal & Garcia, 1968. TYPE: Holotype with
associated larval and pupal skins (Ga 418-17), km 34 of the Vespucio-San Pedri-
to road, near Vespucio (Salta), 6 Oct 1966, Garcia, Hepper and Fernandez (INM).
BIONOMICS: Larvae taken at the margins of an arroyo.

77. Culex (Mel.) misionensis Duret, 1953. TYPE: Holotype & (A.U. 23, E.
14) with genitalia slide, Aristobulo del Valle (Misiones), 28 Feb 1951, Duret
(CMPHM). BIONOMICS: [Larvae probably in permanent or semipermanent
ground waters. |

78. Culex (Mel.) orfilai Duret, 1953. TYPE: Holotype & (D.O 69, E. 20)
with genitalia slide, Iguazu, Villa Tacuara (Misiones), 21 Nov 1951, Duret
(CMPHM). BIONOMICS: [Larvae probably in permanent or semipermanent
ground waters. |

79. Culex (Mel.) pavlovskyi Casal & Garcia, 1967. TYPE: Holotype & with
genitalia slide (660331), Aeropuerto de Camba Punta, near Corrientes (Corrien-
tes), 4 Mar 1966, O.H. Casal (INM). BIONOMICS: Adults taken at a lighted
window. [Larvae probably in permanent or semipermanent ground waters.|

80. Culex oblita (Lynch Arribalzaga, 1891) [NEW COMBINATION; NOMEN
DUBIUM]. TYPE: Holotype “ [only a leg and a wing remain], Navarro, near
Arroyo de las Saladas (Buenos Aires), Mar 1884[F. Lynch] (BA).

81. Sayomyia australis (Shannon & Del Ponte, 1928). TYPE: Lectotype §,
in good condition, with the following labels, ''Ins. Bac. Ent. nota 195/Finca San-
ta Barbara, 6.IV.27 Tuc.[uman, Departamento de la Capital]/R.C. Shannon
coll & Del Ponte/Chaoborus australis Shannon & Del Ponte" (INM; PRESENT
DESIGNATION by O. H. Casal). BIONOMICS: Adults collected at lights. [Lar-
vae probably in permanent or semipermanent still ground waters. | |

82. Sayomyia elnorae (Shannon & Del Ponte, 1928). TYPE: 2, Tres Pozos,
near Embarcacion (Salta), E.S. Shannon (LU). BIONOMICS: Adults collected
at light near a lake. [Larvae probably in lakes and other permanent and semi-
permanent still ground waters. |

83. Corethrella arborealis Shannon & Del Ponte, 1928 [=appendiculata].
TYPE: Holotype o [(174) and wing slide (389)], Vipos [Tucuman], 4 Mar 1927,
E. Del Ponte (INM). BIONOMICS: Holotype bred from pupa from treehole.

84. Corethrella puella Shannon & Del Ponte, 1928. TYPE: Lectotype @, in
good condition, with the following labels, ''Ledesma, Jujuy, 30.3.27/R.C. Shan-
non coll./Corethrella puella S & DP," the last label by Del Ponte (INM; PRES-
ENT DESIGNATION by O. H. Casal). BIONOMICS: Adults collected in bathroom.
[Larvae probably in permanent and semipermanent ground water with dense
vegetation. |

85. Corethrella quadrivittata Shannon & Del Ponte, 1928. TYPE: Lectotype

Belkin et al: Topotypic South American Mosquitoes 17

o", in very good condition, with the following labels, ''Tres Pozos, Salta, 20.4.
27/Shannon & Shannon. /Corethrella quadrivittata Shannon y Del Ponte," the
last label by Shannon (INM; PRESENT DESIGNATION by O. H. Casal). BIO-
NOMICS: Females taken at light near a lake. Larvae in aquatic vegetation (Pis-
tia and others) in lakes.

86. Lutzomiops davisi (Shannon & Del Ponte, 1928). TYPE: Lectotype 9,
in fairly good condition, with the following labels, "Ins. Bac. Ent. nota 32-4/
Concepcion, Tuc.[umdan] 4.7.26/R.C. Shannon/390/Corethrella davisi Shannon
& Del Ponte," the last label by Shannon (INM; PRESENT DESIGNATION by O. H.
Casal). BIONOMICS: Larvae in a ground pool with semistagnant water and with
vegetation at the margins.

87. Nothodixa atrovittata (Edwards, 1930). TYPE: Holotype “, Bariloche
[San Carlos de Bariloche] (Rio Negro), elev. 2450 ft, 25 Oct-1 Dec 1926, F. W.
Edwards (BM). BIONOMICS: [Larvae probably in marginal vegetation or flo-
tage in small creeks. |

88. Nothodixa nitida (Edwards, 1930). TYPE: Holotype 0, Bariloche [San
Carlos de Bariloche] (Rio Negro), elev. 2450 ft, 25 Oct-1 Dec 1926, F.W. Ed-
wards (BM). BIONOMICS: [Larvae probably in marginal vegetation or flotage
in small creeks.|

89. Dixella argentina (Alexander, 1920). TYPE: Holotype 0, La Granja,
Alta Gracia (Cordoba), 1-8 Apr 1920, Charles Bruch(A). BIONOMICS: [Larvae
probably in various types of permanent and semipermanent ground waters, es-
pecially at margins of small streams. |

List of Localities

BUENOS AIRES

[Locality not specified]: 38. Psorophora (P.) lynchi; 44. Psorophora (J.)
bruchi; 48. Aedes (O.) lynchii.

Baradero: 1. Anopheles (A.) annulipalpis; 6. Anopheles (N.) albitarsis; 37.
Aedeomyia squamipennis; 63. Culex (C.) dolosus.

Chacabuco: 63. Culex (C.) dolosus.

Las Conchas: 63. Culex (C.) dolosus.

Lujan (Rio): 34. Uranotaenia pulcherrima.

Navarro: 28. Coquillettidia (R.) fasciolata; 54. Aedes (O.) confirmatus
(banks of Rio Salado); 63. Culex (C.) dolosus.

Otamendi (Ingeniero Romulo) (Canal 6 and Parana de las Palmas, Estacion
Agropecuaria Delta de I. N. T. A.): 62. Culex (C.) castroi; 65. Culex (C.) eduar-
doi; Culex (C.) hepperi.

Parana, Islas del (Tuyuparé, a brook in the delta of the Parana): 18. Trich-
oprosopon (R.) paranensis; 22. Wyeomyia (M.) leontiniae; 42. Psorophora (J.)
centrale.

San Isidro: 49. Aedes (O.) tapinops; 61. Culex (C.) brethesi; 64. Culex (C.)
bonariensis; 75. Culex (Mel.) intrincatus (lowlands).

Tigre: 32. Uranotaenia nataliae (Rio Las Conchas).

Zarate: 63. Culex (C.) dolosus.

CHACO |
[Locality not specified]: 25. Phoniomyia mueblensi.
Resistencia: 31. Uranotaenia monilis; 35. Uranotaenia urania.

18 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

CORDOBA
Achiras: 68. Culex (C.) lahillei.
La Granja, Alta Gracia: 89. Dixella wiventina.
Primero (Rio): 70. Culex (C.) autumnalis.

CORRIENTES
[Locality not specified]: 10. Anopheles (N.) bachmanni.
Corrientes: 79. Culex (Mel.) pavlovskyi (Aeropuerto de Camba Punta).

ENTRE RIOS
[Locality not specified]: 10. Anopheles (N.) bachmanni.

FORMOSA

[Locality not specified]: 10. Anopheles (N.) bachmanni; 16. Toxorhynchites
(L.) separatus; 17. Toxorhynchites (L.) lynchi; 30. Uranotaenia lanei; 39. Psor-
ophera (P.) holmbergii; 45. Psorophora (G.) confinnis.

San Jose: 21. Wyeomyia (Dav.) monoleua.

J UJUY

[Locality not specified]: 56. Haemagogus (S.) spegazzinii; 60. Culex (C.) in-
terfor (on train between Tucuman and Jujuy).

Ledesma: 24. Wyeomyia (D.) typharum; 29. Coquillettidia (R.) araozi; 33.
Uranotaenia capitis; 84. Corethrella puella (Cafiitas Viejo).

San Pedro: 52. Aedes (O.) patersoni; 53. Aedes (O.) raymondi; 58. Culex
(Lutzia) patersoni; 69. Culex (C.) maxi. —

Santa Clara: 27. Sabethes (S.) neivai.

Zapla: 19. Wyeomyia (N.) lateralis.

MENDOZA
Monte Coman: 72. Culex (All.) tramazayguesi.

MISIONES

[Locality not specified, probably Cataratas del Iguazu]: 50. Aedes (O.)
iguazu.

Aristobulo del Valle: 77. Culex (Mel.) misionensis.

Bernardo de Irigoyen: 74. Culex (Mel.) bejaranoi.

Cataratas: 23. Wyeomyia (D.) belkini (20 km west of).

Cerro Azul, Arroyo San Juan: 73. Culex (Mel.) aliciae.

Corpus, San Ignacio: 2. Anopheles (A.) holmbergi.

Iguazu (Cataratas del): 26. Limatus exhibitor.

Iguazu, Villa Tacuara: 78. Culex (Mel.) orfilai.

Puerto Iguazu: 13. Toxorhynchites (L.) cavalierii (50m from Hosteria Ruf-
fino).

San Pedro: 59. Culex (C.) ameliae (10 km west of, on San Pedro-El Dorado
road).

RIO NEGRO
San Carlos de Bariloche, 2450 ft: 87. Nothodixa atrovittata; 88. Nothodixa
nitida.

SALTA
[Locality not specified]: 60. Culex (C.) interfor (on train between Tucuman
and Jujuy).

Belkin et al: Topotypic South American Mosquitoes 1?

Campo Santo de Salta: 71. Culex (C.) saltanensis.

Piquirenda: 66. Culex (C.) fernandezi.

Reserva Nacional "Finca del Rey," Departamento de Anta: 47. Aedes (O.)
meprai.

Salvador Mazza (Pocitos), Departamento General San Martin: 55. Haema-

ogus (S.) petrocchiae.
San Lorenzo (Quebrada de), near Salta: 51. Aedes (O.) araozi.

Tres Pozos, near Embarcacidn: 12. Anopheles (N.) davisi; 36. Uranotaenia

ere

elnora; 43. Psorophora (J.) chaquensis; 46. Psorophora (G.) paulli; 82. Sayomy-
ia elnorae; 85. Corethrella quadrivittata.

Vespucio: 76. Culex (Mel.) martinezi (km 34 of Vespucio-San Pedrito road).

SANTA FE
Tapenaga (Rio), Colonia Florencia, Gran Chaco: 40. Psorophora (P.) agog-

gylia; 41. Psorophora (P.) stigmatephora.

TUCUMAN

[Locality not specified]: 3. Anopheles (A.) argentinus; 5. Anopheles (A.)
patersoni; 7. Anopheles (N.) rooti; 8. Anopheles (N.) evansae; 14. Toxorhynchi-
tes (L.) tucumanus; 60. Culex (C.) interfor (on train between Tucuman and Ju-
juy).

Concepcion: 86. Lutzomiops davisi (on road 5 km W).

Lules: 15. Toxorhynchites (L.) arborealis.
Monteros: 9. Anopheles (N.) clarki.

Raco: 20. Wyeomyia (Dav.) petrocchiae.
Salf (Rio): 4. Anopheles (A.) tucumanus.
Santa Barbara (Finca), Departamento dela Capital: 11. Anopheles (N.) per-

ezi; 81. Sayomyia australis.
Vipos: 57. Haemagogus (S.) uriartei; 83. Corethrella arborealis.

BOLIVIA

List of Species

1. Anopheles (Kerteszia) boliviensis Theobald, 1905. TYPE: Holotype e,
Songo [Zongo] (La Paz), date not specified, M. Bird (BM). BIONOMICS: |Lar-
vae probably in leaf axils of bromeliads. |

2. Toxorhynchites (Ankylorhynchus) hexacis (Martini, 1931). TYPE: Holo-
type 9, Yungas de Coroico, Nor Yungas (La Paz), elev. 1000 m (NE). BIONOM-
ICS: [Larvae probably in treeholes or leaf axils.|

3. Trichoprosopon (Ctenogoeldia) magnum (Theobald, 1905). TYPE: Holo-
type ?, San Antonio [de Mapiri] (La Paz), M. Biro (HNM). BIONOMICS: [Lar-
vae probably in leaf axils and flower bracts of Calathea or other Marantaceae. |

4. Mansonia (M.) fonsecai (Pinto, 1932) [=indubitans]|. TYPE: Holotype ?
(160), Los Naranjos [Naranjo] (Santa Cruz), 23 Feb 1925, O. da Fonseca (IOC).
BIONOMICS: [Larvae probably attached to floating vegetation (Pistia) in ponds
or lakes.|

5. Coquillettidia (Rhynchotaenia) nitens (Cerqueira, 1943). TYPE: Holo-
type 2, Puerto Suarez, Chiquitos (Santa Cruz), July 1939 (IOC). BIONOMICS:
[Larvae probably on rootlets of grasses and other herbaceous vegetation in mud
or fine sediment in very shallow water on margins of swamps, ponds or lakes.|

20 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

6. Orthopodomyia bacigalupoi Martinez & Prosen, 1958 [=fascipes]. TYPE:
Holotype o, Tact, Buenavista, Ichilo (Santa Cruz), Apr 1955, A. Martinez and

A. F. Prosen (BA). BIONOMICS: Larvae in treeholes.

7. Psorophora (Janthinosoma) circumflava Cerqueira, 1943. TYPE: Holo-
type °, Axiamas, Caupolican (La Paz), Dec 1941 (IOC). BIONOMICS: | Larvae
probably in temporary ground pools or overflows of ponds or streams. |

8. Psorophora (J.) melanota Cerqueira, 1943. TYPE: Holotype <, Riber-
alta, Terr. de Colonias (El Beni), Dec 1939 (IOC). BIONOMICS: [Larvae prob-
ably in temporary ground pools or overflows of ponds or streams.|

9. Psorophora (Grabhamia) chiquitana Pinto, 1932 [=confinnis]. TYPE:
Holotype 2 (163), Los Naranjos [Naranjo] (Santa Cruz), 1925, O. da Fonseca
(IOC). BIONOMICS: [Larvae probably in temporary ground pools or overflows
of ponds or streams.|

10. Psorophora (G.) dimidiata Cerqueira, 1943. TYPE: Holotype“, La-
eunillas, Valle Grande (Santa Cruz), Feb 1940 (IOC). BIONOMICS: [Larvae
probably in temporary ground pools or overflows of ponds or streams. |

11. Aedes (Ochlerotatus) bejaranoi Martinez, Carcavallo & Prosen, 1960.
TYPE: Holotype o°, Yungas del Palmar, between km 90 and 110 on road from
Cochabamba to Villa Tunari [San Antonio), Chapare (Cochabamba), elev. 3000-
3900 m, Feb 1959, A. Martinez (BA). BIONOMICS: Larvae in pools of very
cold, slowly flowing water and with dense vegetation and abundant filamentous
algae. Adults of both sexes resting in vegetation on margins of pools.

12. Aedes (O.) oroecetor Martini, 1931 miter TYPE: Lectotype
with associated genitalia slide, Sorata (La Paz), elev. 2300 m (BM; designation
of Mattingly, 1955:29). BIONOMICS: [Larvae probably in shaded stream bed
pools with organic debris and surrounded by vegetation. |

13. Aedes (Howardina) aurivittatus Cerqueira, 1943. TYPE: Holotype 9,
Pampa Grande, Florida (Santa Cruz), Feb 1940 (IOC). BIONOMICS: [Larvae
probably in leaf axils of bromeliads, possibly in treeholes.|

14. Aedes (H.) delpontei Martinez & Prosen, 1955. TYPE: Holotype , be-
tween km 125 (Limbo, elev. 2000 m) and km 145 (Locotal, elev. 1200 m) on road
from Cochabamba to Villa Tunari [San Antonio], Chapare (Cochabamba), Jan
1949, Feb 1952 or Nov 1953, A. Martinez and A. F. Prosen (BA). BIONOMICS:
[Larvae probably in leaf axils of bromeliads, possibly in treeholes. |

15. Aedes (H.) vanemdeni Martini, 1931. TYPE: Lectotype o, Yungas de
Coroico, Nor Yungas (La Paz), elev. 1000 m (BM; designation of Mattingly,
1955:31). BIONOMICS: [Larvae probably in leaf axils of bromeliads, possibly
in treeholes.|

16. Haemagogus (Stegoconops) lindneri Martini, 1931 [=speggazzinii or uri-
artei|. TYPE: Holotype ?, [?San José de] Chiquitos (Santa Cruz), Oct 1926
1926 (SMNS). BIONOMICS: [Larvae probably in treeholes. |

17. Culex (C.) alticola Martini, 1931. TYPE: Lectotype 0, Serata [Sorata]
(La Paz), elev. 2300 m, 19 Dec 1902 (BM; designation of Mattingly, 1955:31).
BIONOMICS: [Larvae probably in permanent or semipermanent ground waters. |

18. Culex (C.) bidens Dyar, 1922. TYPE: Lectotype & with genitalia slide
(1687), Rosario, near Lago Rogagua (El Beni), Nov 1921, W.M. Mann (USNM,
25760; designation of Stone and Knight, 1957a:47). BIONOMICS: [Larvae prob-
ably in permanent and semipermanent ground waters.| Adults probably collec-
ted in forest islands in generally open country.

19. Edwardsops boliviensis Lane & Heredia, 1956. TYPE: Holotype ¢, El
Paylon [E1 Pailon| (Santa Cruz), 5-6 Feb 1955, Travassos, Barros and Albu-
querque (FH). BIONOMICS: [Larvae probably in permanent or semipermanent
ground pools, possibly in treeholes. |

Belkin et al: Topotypic South American Mosquitoes 21

List of Localities

BENI (EL)
Riberalta: 8. Psorophora (J.) melanota.
Rosario, near Lago Rogaguaz 18. Culex (C.) bidens.
San Antonio: (See under LA PAZ).

COCHABAMBA |

Chapare Prov.; road from Cochabamba to Villa Tunari (San Antonio): 11.
Aedes (O.) bejaranoi (Yungas del Palmar, between km 90 and 110; elev. 3000-
3900 m); 14. Aedes (H.) delpontei (between km 125, elev. 2000 m, and km 145,
elev. 1200 m). :

LA PAZ

Axiamas, Caupolican Prov.: 7. Psorophora (J.) circumflava.

Coroico (Yungas de) elev. 1000 m, Nor Yungas Prov.: 2. Toxorhynchites
(A.) hexacis; 15. Aedes (H.) vanemdeni.

San Antonio de Mapiri: 3. Trichoprosopon (Ctenogoeldia) magnum.

Sorata: 12. Aedes (O.) oroecetor; 17. Culex (C.) alticola.

Zongo (Songo): 1. Anopheles (K.) boliviensis.

SANTA CRUZ

Buenavista (Tact), Ichilo Prov.: 6. Orthopodomyia bacigalupoi.

Chiquitos (?San José de), Agua retirada, Chiquitos Prov.: 16. Haemagogus
(S.) lindneri.

Lagunillas, Valle Grande Prov.: 6. Psorophora (G.) dimidiata.

Naranjo (Los Naranjos): 4. Mansonia(M.) fonsecai; 9. Psorophora (G.) chi-
quitana.

Pailon (El): 19. Edwardsops boliviensis.

Pampa Grande, Florida Prov.: 13. Aedes (H.) aurivittatus.

Puerto Suarez, Chiquitos Prov.: 5. Mansonia nitens.

CHILE

List of Species

1. Anopheles (A.) neghmei Mann, 1950 [ssp. of pseudopunctipennis|. TYPE:
Holotype o, Quebrada de Mifiemifie (Tarapaca), elev. 1800 m, Feb-Mar or
June 1946, G. Mann F. (Inst. Biol. "Juan Noe,"' Santiago, A-1). BIONOMICS:
Larvae in natural and artificial ground waters.

2. Anopheles (A.) noei Mann, 1950 [ssp. of pseudopunctipennis]. TYPE:
Holotype o", Oasis of Suca (Tarapaca), elev. 1330 m, Feb-Mar or June 1946,
G. Mann F. (Inst. Biol. 'Juan Noe,'' Santiago, A-32). BIONOMICS: Larvae
primarily in irrigation waters.

3. Anopheles (Nyssorhynchus) pictipennis Philippi, 1865. TYPE: 20; San-
tiago (Santiago), July 1859; Aconcagua province, Sept 1863 (NE). BIONOM-
ICS: [Larvae probably in permanent or semipermanent ground waters. |

4. Anopheles (N.) bigotii Theobald, 1901 [=pictipennis]. TYPE: Holotype
°, locality not specified, 1894, J.M.F. Bigot (NE). BIONOMICS: [Larvae prob-
ably in permanent or semipermanent ground waters. |

5. Anopheles (N.) variegatus (E. Blanchard, 1852); chilensis (R. Blan-

oe Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

chard, 1905), nom.nov. [=pictipennis]. TYPE: Lectotype ?, Arquero [Arque-
ros?] (MNHP; designation of Belkin, 1968:10). BIONOMICS: [Larvae probably
in permanent or semipermanent ground waters. |

6. Psorophora marmorata (Philippi, 1865) [identity doubtful]. TYPE: 2 ¢,
1°, locality not specified (NE). BIONOMICS: [Larvae probably in ground pools. |

7. Aedes (Ochlerotatus) flavipes (Macquart, 1838)[=?albifasciatus]. TYPE:
Holotype ?, Concepcion (Concepcion), no other data (MNHP). BIONOMICS: [Lar-
vae probably in temporary ground pools or flooded margins of ponds or streams. |

8. Aedes (O.) annuliferus (E. Blanchard, 1852) [=?albifasciatus]. TYPE:
Lectotype 2, probably from vicinity of Coquimbo or Illapel (Coquimbo) (MNHP;
designation of Belkin, 1968:4). BIONOMICS: [Larvae probably in temporary
ground pools or flooded margins of ponds or streams.| _

9. Aedes (O.) vittatus (Philippi, 1865); philipii Dyar, 1924, new name [=
?albifasciatus]|. TYPE: 2 9, Santiago (Santiago), no other data (NE). BIONOM-
ICS: [Larvae probably in temporary ground pools or flooded margins of ponds
or streams. |

10. Aedes (O.) colonarius Dyar, 1924 [=?albifasciatus]. TYPE: Holotype
?, Azapa Valley (Tarapaca), June 1912, C.E. Porter (USNM). BIONOMICS:
[Larvae probably in temporary ground pools or flooded margins of streams. |

11. Culex (C.) annuliventris (E. Blanchard, 1852) [A distinct species].
TYPE: Lectotype &, possibly from Valdivia (Valdivia) (MNHP; designation of
Belkin, 1968:13). BIONOMICS: [Larvae in permanent or semipermanent ground
waters. |

12. Culex (C.) apicinus Philippi, 1865. TYPE: Adult, near Santiago (San-
tiago), no other data (NE). BIONOMICS: [Larvae probably in permanent or sem-
ipermanent ground waters, ponds or streams.|

13. Culex (C.) articularis Philippi, 1865. TYPE: Original material from
near Corral (Valdivia) all lost (NE); neotype o’, Casa Pangue (Llanquihue), Dec
1926, R. and E. Shannon (USNM, RB 62 #554; designation of Bram, 1967:30).
BIONOMICS: [Larvae in permanent or semipermanent ground waters. |

14. Culex (C.) serotinus Philippi, 1865 [=pipiens complex]. TYPE: , 9,
Santiago (Santiago) and Valdivia (Valdivia), no other data (NE). BIONOMICS:
[Larvae probably in ground waters contaminated with domestic wastes or in
large artificial containers. |

15. Nothodixa chilensis (Alexander, 1913). TYPE: Holotype o, Concepcion
(Concepcion), 23 Aug 1904, P. Herbst (NE). BIONOMICS: [Larvae probably in
vegetation or flotage on margins of streams, rivulets, seepages and springs. |

16. Nothodixa ensifera (Edwards, 1930). TYPE: Holotype “, Casa Pangue
(Llanquihue), elev. about 1000 ft, 4-10 Dec 1926, F.W. Edwards (BM). BIO-
NOMICS: [Larvae probably in vegetation or flotage on margins of streams, riv-
ulets, seepages and springs. |

List of Localities

[NOT SPECIFIED]
Locality and province not specified, probably near Santiago or Valparaiso:

4. Anopheles (N.) bigotii; 6. Psorophora marmorata.

ACONCAGUA
[Locality not specified]: 3. Anopheles (N.) pictipennis (also Santiago).

Belkin et al: Topotypic South American Mosquitoes 23

CONCEPCION
Concepcion: 7. Aedes (O.) flavipes; 15. Nothodixa chilensis.

COQUIMBO
Arqueros: 5. Anopheles (N.) variegatus.
Coquimbo: 8. Aedes (O.) annuliferus (also Illapel).
Illapel: 8. Aedes (O.) annuliferus (also Coquimbo).

LLANQUIHUE
Casa Pangue, frontier station on Chilean side of Rosales Pass, altitude of
house about 1000 ft: 13. Culex (C.) articularis; 16. Nothodixa ensifera.

SANTIAGO

Santiago: 3. Anopheles (N.) pictipennis (also Aconcagua Prov.); 9. Aedes
(O.) vittatus; 14. Culex (C.) serotinus (also Valdivia).

See also [NOT SPECIFIED].

TARAPACA
Azapa Valley: 10. Aedes (O.) colonarius.
Mifiemifie, Quebrada de: 1. Anopheles (A.) neghmei.
Suca, Oasis of: 2. Anopheles (A.) noei.

VALDIVIA

Corral: 13. Culex (C.) articularis.

Valdivia: 11. Culex (C.) annuliventris; 14. Culex (C.) serotinus (also San-
tiago).

PARAGUAY

_ List of Species

1. Phoniomyia fuscipes (Edwards, 1922). TYPE: Lectotype ?, unspecified
locality in Paraguay, probably Asuncion (Asuncion), Fiebrig (BM; designation
of Belkin, 1968:24). BIONOMICS: [Larvae probably in leaf axils of bromeliads. ]

2. Psorophora (P.) pallescens Edwards, 1922. TYPE: Lectotype % with
attached genitalia mount, unspecified locality in Paraguay, probably Asuncion
(Asuncion), Fiebrig (BM: designation of Belkin, 1968:28). BIONOMICS: [Lar-
vae probably in temporary ground pools or overflows of ponds and streams.|

3. Psorophora (Janthinosoma) purpurascens Edwards, 1922 [=cyanescens].
TYPE: Holotype, locality and date not specified, probably Asuncion (Asuncion),
Fiebrig (HNM). BIONOMICS: [Larvae probably in sunlit temporary ground
pools. |

4. Psorophora (J.) fiebrigi Edwards, 1922 [=ferox]. TYPE: Lectotype &
with attached genitalia mount, unspecified locality in Paraguay, probably Asun-
cion (Asuncion), Fiebrig (BM; designation of Belkin, 1968:26). BIONOMICS:
[Larvae probably in temporary ground pools in wooded or shaded areas.]|

5. Psorophora (J.) paraguayensis (Strickland, 1911) [=varipes]. TYPE:
Holotype 2, Puerto Max (Boqueron), Jan-Apr 1905, Vezényi (NE). BIONOMICS:
[Larvae probably in temporary pools in wooded or shaded areas. |

6. Psorophora (Grabhamia) varinervis Edwards, 1922. TYPE: Holotype 9,
locality and date not specified, probably Asuncion (Asuncion), Fiebrig (HNM).

24 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

BIONOMICS: [Larvae probably in temporary ground pools or overflows of ponds
and streams.|

7. Aedes (Ochlerotatus) stigmaticus Edwards, 1922. TYPE: Lectotype ¢,
Asuncion (Asuncion), 1904, Vezényi (BM; designation of Belkin, 1968:7). BIO-
NOMICS: [Larvae probably in temporary ground pools or overflows of ponds
and streams.|

List of Localities

ASUNCION

[Locality not specified, probably vicinity of Asuncion]: 1. Phoniomyia fus-
cipes; 2. Psorophora (P.) pallescens; 3. Psorophora (J.) purpurascens; 4. Psor-
ophora (J.) fiebrigi; 6. Psorophora (G.) varinervis.

Asuncion: 7. Aedes (O.) stigmaticus.

BOQUERON
Puerto Max: 5. Psorophora (J.) paraguayensis.

PERU

List of Species

1. Anopheles (Stethomyia) acanthotorynus Komp, 1937. TYPE: Holotype &
with genitalia slide, Iquitos (Loreto), Apr 1931, R.C. Shannon (USNM, 52020).

BIONOMICS: [Larvae probably in deeply shaded stream bed pools.|

2. Anopheles (A.) peruvianus Tamayo, 1907 [=pseudopunctipennis]. TYPE:
o', 2, Huacachina (Ica), San Pedro de Lloc (La Libertad), Lima, summer and
fall, Chanchamayo (LU). BIONOMICS: [Larvae probably in various small, sun-
lit, permanent and semipermanent ground waters with algae.|

3. Trichoprosopon (T.) subsplendens (Martini, 1931) [=digitatum]. TYPE:
Holotype 2, Pt. Bermudas [Bermudez], Pichis [Rio] (Pasco), 10/12.03 (NE).
BIONOMICS: [Larvae probably in treeholes, cut bamboo or plant parts on the
ground. |

4. Trichoprosopon (Runchomyia) hyperleucum (Martini, 1931). TYPE: Ho-
lotype 2, Urubamba River (Cuzco) (NE). BIONOMICS: [Larvae probably in tree-
holes, cut bamboo, plant parts on the ground, or in the leaf axils and/or flower
bracts of Araceae, Marantaceae, Musaceae or bromeliads. |

5. Aedes (Ochlerotatus) epinolus Dyar & Knab, 1914 [=taeniorhynchus].
TYPE: Lectotype ?, Ventanillas (Cajamarca), 4 Feb 1914, C.H.T. Townsend
(USNM, 18362; designation of Stone and Knight, 1956a:217). BIONOMICS: [Lar-
vae probably in coastal saltmarshes or freshwater pools near the sea.|

6. Haemagogus (Stegoconops) obscurescens Martini, 1931 [=anastasionis].
TYPE: Lectotype 9, Ucayali River (Loreto) (BM; designation of Mattingly, 1955:
28). BIONOMICS: [Larvae probably in treeholes or cut bamboo. |

7. Culex (C.) debilis (Dyar & Knab, 1914) [=apicinus]. TYPE: Lectotype
o' with genitalia and 3 legs on slides (597, 598), Matucan [Matucana] (Lima),
elev. 7300 ft, June-July 1913, C.T. Brues (USNM, 18361; designation of Stone
and Knight, 1957a:47). _BIONOMICS: [Larvae probably in permanent or semi-
permanent ground waters. |

8. Culex (C.) escomeli Bréthes, 1920 [=apicinus]. TYPE: o, ?, Arequipa
(BA). BIONOMICS: [Probably as for 7. debilis.

Belkin et al: Topotypic South American Mosquitoes 20

9. Culex (C.) archegus Dyar, 1929. TYPE: Holotype o, Tarma (Junin),
elev. 10,000 ft, 13 July 1928, R.C. Shannon (USNM). BIONOMICS: [Larvae prob-
ably in permanent or semipermanent ground waters. |

10. Culex (C.) diplophyllum Dyar, 1929. TYPE: Lectotype o with genitalia
slide (2398), Verrugas Canyon (Lima), 5 Apr 1928, R.C. Shannon (USNM; des-
ignation of Stone and Knight, 1957a:47). BIONOMICS: [Larvae probably in per-
manent or semipermanent ground waters. |

11. Culex (C.) raymondii Tamayo, 1907 [=pipiens quinquefasciatus]. TYPE:
o, 2, larva, pupa, Huacachina (Ica) (LU). BIONOMICS: [Larvae probably in
ground waters contaminated with domestic wastes or in large artificial contain-
ers. |

12. Corethrella inca Lane, 1939 [=ananacola]. TYPE: Holotype ?, Iquitos
(Loreto), Mar-Apr 1931, R.C. Shannon (FH). BIONOMICS: [Larvae probably
in leaf axils of bromeliads, possibly Ananas magdalenae. |

13. Corethrella maculata Lane, 1939. TYPE: Holotype o, Iquitos (Loreto),
Mar-Apr 1931, R.C. Shannon (FH). BIONOMICS: [Larvae probably in ground
waters, possibly in leaf axils or treeholes.|

14. Lutzomiops peruviana (Lane, 1939). TYPE: Holotype ?, Iquitos (Lor-
eto), Mar-Apr 1931, R.C. Shannon (FH). BIONOMICS: [Larvae probably in
swamps. | 7

15. Lutzomiops shannoni (Lane, 1939). TYPE: Holotype o, Iquitos (Lore-
to), Mar-Apr 1931, R.C. Shannon (USNM). BIONOMICS: Larvae in a swamp.

16. Dixella andeana (Lane, 1942). TYPE: Holotype o, Iquitos (Loreto),
Mar-Apr 1931, R.C. Shannon (USNM). BIONOMICS: [Larvae probably on mar-
gins of small streams or permanent or semipermanent ground pools. |

17. Dixella peruviana (Edwards, 1931). TYPE: Holotype “, Verrugas
(Lima), 5 May 1928, R.C. Shannon (BM). BIONOMICS: [Larvae probably on
margins of small streams or springs. |

List of Localities

AREQUIPA
[Locality not specified]: 8. Culex (C.) escomeli.

CAJAMARCA
Ventanillas: 5. Aedes (O.) epinolus.

CUZCO
Urubamba (Rio): 4. Trichoprosopon (R.) hyperleucum.

ICA
Huacachina: 2. Anopheles (A.) peruvianus (also San Pedro de Lloc (La Li-
bertad) and Lima); 11. Culex (C.) raymondii.

J UNIN
Tarma: 9. Culex (C.) archegus.

LA LIBERTAD
San Pedro de Lloc: 2. Anopheles (A.) peruvianus (also Huacachina (Ica) and
Lima).

26 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

LIMA

[Locality not specified, probably vicinity of Lima]: 2. Anopheles (A.) peru-
vians (also Huacachina (Ica) and San Pedro de Lloc, La Libertad).

Matucuna: 7. Culex (C.) debilis.

Verrugas: 10. Culex (C.) diplophyllum (Verrugas Canyon); 17. Dixella pe-
ruviana.

LORETO

Iquitos: 1. Anopheles (S.) acanthotorynus; 12. Corethrella inca; 13. Core-
thrella maculata; 14. Lutzomiops peruviana; 15. Lutzomiops shannoni; 16. Dix-
ella andeana.

Ucayali (Rio): 6. Haemagogus (S.) obscurescens.

PASCO !
Puerto Bermudez, Rio Pichis: 3. Trichoprosopon (T.) subsplendens.

URUGUAY

List of Species

1. Psorophora (Grabhamia) leucocnemis Martini, 1931 [?=varinervis].
TYPE: Holotype 9, Montevideo (Montevideo), 19 Mar 1927, Vogelsang coll. 8846

(BM). BIONOMICS: [Larvae probably in temporary ground pools or overflows
of ponds or streams. |

REFERENCES CITED

Belkin, John N.

1968. Mosquito Studies (Diptera, Culicidae). IX. The tbe specimens of
New World mosquitoes in European museums. Amer. Entomol. Inst.,
Contrib. 3(4). 69 p.

Belkin, John N., R. X. Schick and $.J. Heinemann

1965. Mosquito Studies (Diptera, Culicidae). V. Mosquitoes originally de-
scribed from Middle America. Amer. Entomol. Inst., Contrib. 1(5):
1-99,

1966. Mosquito Studies (Diptera, Culicidae). VI. Mosquitoes originally de-
scribed from North America. Amer. Entomol. Inst., Contrib. 1(6):
1-39.

Bram, Ralph A.

1967. Classification of Culex subgenus Culex in the New World. U.S. Nat.

Mus., Proc. 120:1-122.
Castro, Manuel and M. Bressanello

1952. Revision de las especies de Taeniorhynchus (Rhyncotaenia) (Dipt.

Cul.). Rev. Brasil. Biol., Rio de Janeiro 12:229-246.
Garcia, Miguel and R.A. Ronderos

(1962) 1963. Sobre la validez de Aedes (Ochlerotatus) patersoni Shannon y
del Ponte, 1927 (Diptera-Culicidae). Soc. Entomol. Argent., Rev.,
Buenos Aires 25:29-33.

Peters, T. Michael
1968. Mosquito Studies (Diptera, Culicidae). X. Dixinae originally described

Belkin et al: Topotypic South American Mosquitoes at

from North America. Amer. Entomol. Inst., Contrib. 4(1):1-7.
Stone, Alan and K. L. Knight :

1955. Type specimens of mosquitoes in the United States National Museum:
I. The genera Armigeres, Psorophora and Haemagogus (Diptera, Cu-
licidae). Wash. Acad. Sci., J. 45:282-289.

1956. Type specimens of mosquitoes in the United States National Museum:
III. The genera Anopheles and Chagasia (Diptera, Culicidae). Wash.
Acad. Sci., J. 46:276-280.

1957a. Type specimens of mosquitoes in the United States National Muse-
um: IV. The genus Culex (Diptera, Culicidae). Wash. Acad. Sci., J.
47:117-126.

1957b. Type specimens of mosquitoes in the United States National Muse-
um: V. The Sabethini (Diptera, Culicidae). Wash. Acad. Sci., J. 47:
117-120. |

1957c. Type specimens of mosquitoes in the United States National Muse-
um: VI, Miscellaneous genera, addenda, and summary. Wash. Acad.
Sci., J. 47:196-202. |

Umana, A.C., R.L. Heredia and J.C. Siquot

1959. Estudios sobre el Anopheles (N.) albitarsis en la Argentina (Nota
previa). Primeras Jornadas Entomoepidemiologicas Argentinas,
Buenos Aires, part 2, p. 609-618.

CLASSIFIED INDEX

AEDEOMYIA [1] stigmaticus, 24

*squamipennis, 12

AEDES (HOWARDINA) [3]

aurivittatus, 20
delpontei, 20
vanemdeni, 20

AEDES (OCHLEROTATUS) [16]

albifasciatus
annuliferus, 22
colonarius, 22
flavipes, 22
#philipii, 22
vittatus, 22
angustivittatus
meprai, 13
bejaranoi, 20
crinifer
iguazu, 14
lynchii, 13
tapinops, 13
milleri
araozi, 14
oroecetor, 20
patersoni, 14
raymondi, 14

scapularis

*confirmatus, 14

taeniorhynchus
epinolus, 24
ANOPHELES (ANOPHELES) [8]
annulipalpis, 9
fluminensis
holmbergi, 9
pseudopunctipennis
*argentinus, 9
neghmei, 21
noei, 21
patersoni, 10
peruvianus, 24
tucumanus, 10
ANOPHELES (KERTESZIA) [1]
boliviensis, 19
ANOPHELES (NYSSORHYNCHUS) [10]
*albitarsis, 10
argyritarsis
rooti, 10
evansae, 10
noroestensis
clarki, 10
pictipennis, 21
bigotii, 21
#chilensis, 21
variegatus, 21

28 Contrib. Amer. Ent. Inst., vol. 4, no. 1, 1968

triannulatus
*bachmanni, 10
davisi, 10
perezi, 10
ANOPHELES (STETHOMYIA) [1]
acanthotorynus, 24

bejaranoi, 16
intrincatus, 16
martinezi, 16
misionensis, 16
orfilai, 16
pavlovskyi, 16

COQUILLETTIDIA (RHYNCHOTAENIA)| CULEX (subgenus unknown) [1]

[3]
fasciolata, 12
nitens, 19
venezuelensis
araozi, 12
CORETHRELLA [5]
ananacola
inca, 25
appendiculata
arborealis, 16
maculata, 25
puella, 16
quadrivittata, 16
CULEX (ALLIMANTA) [1]
*tramazayguesi, 15
CULEX (CULEX) [23]
ameliae, 14
annuliventris, 22
apicinus, 22
debilis, 24
escomeli, 24
archegus, 20
articularis, 22
bidens, 20
interfor, 14
brethesi, 15
castroi, 15
diplophyllum, 25
*dolosus, 15
bonariensis, 15
eduardoi, 15
fernandezi, 15
hepperi, 15
lahillei, 15
maxi, 15
pipiens
autumnalis, 15
raymondii, 25
serotinus, 22
saltanensis, 15
CULEX (LUTZIA) [1]
bigoti
patersoni, 14
CULEX (MELANOCONION) [7]
aliciae, 16

oblita, 16
DIXELLA [3]
andeana, 25
argentina, 17
peruviana, 25
EDWARDSOPS [1]
boliviensis, 20
HAEMAGOGUS (STEGOCONOPS) [5]
anastasionis
obscurescens, 24
capricornii
petrocchiae, 14
spegazzinii, 14
lindneri, 20
uriartei, 14
LIMATUS [1]
durhamii
exhibitor, 11
LUTZOMIOPS [3]
davisi, 17
peruviana, 25
shannoni, 25

/MANSONIA (MANSONIA) [1]

indubitans
fonsecai, 19
NOTHODIXA [4]
atrovittata, 17
chilensis, 22
ensifera, 22
nitida, 17
ORTHOPODOMYIA |[1}
fascipes
bacigalupoi, 20
PHONIOMYIA [2]
fuscipes, 23
muehlensi, 11 i
PSOROPHORA (GRABHAMIA) [6]
*confinnis, 13
chiquitana, 20
dimidiata, 20
paulli, 13
varinervis, 23
leucocnemis, 26
PSOROPHORA (JANTHINOSOMA) [8]
circumflava, 20

Belkin et al: Topotypic South American Mosquitoes

cyanescens

purpurascens, 23
ferox
centrale, 13
fiebrigi, 23
lutzii
chaquensis, 13
melanota, 20 —
varipes
bruchi, 13.
paraguayensis, 23
PSOROPHORA (PSOROPHORA) [5]
ciliata
lynchi, 12
holmbergii, 13
agoggylia, 13
pallescens, 23
stigmatephora, 13
PHOROPHORA (subgenus unknown) [1]
marmorata, 22
SABETHES (SABETHES) [1]
albiprivus
neivai, 12
SAYOMYIA [2]
australis, 16
elnorae, 16
TOXORHYNCHITES (ANKYLORHYN-
CHUS) [1]
hexacis, 19
TOXORHYNCHITES (LYNCHIELLA) [5]
cavalierii, 10
guadeloupensis
arborealis, 10
tucumanus, 10

29

haemorrhoidalis
lynchi, 11
separatus, 11
TRICHOPROSOPON (CTENOGOELDIA)
1]
magnum, 19
TRIC HOPROSOPON (RUNCHOMYIA) [2]
hyperleucum, 24
paranensis, 11
TRICHOPROSOPON (TRICHOPROSO-
PON) [1]
digitatum
subsplendens, 24
URANOTAENIA [77]
capitis, 12
lanei, 12
monilis, 12
*nataliae, 12
*pulcherrima, 12
elnora, 12
urania, 12
WYEOMYIA (DAVISMYIA) [2]
*petrocchiae, 11
monoleua 11
WYEOMYIA (DENDROMYIA) [2]
belkini, 11 :
melanocephala
typharum, I1
WYEOMYIA (MENOLEPIS) [1]
leucostigma
leontiniae, 11
WYEOMYIA (NUNEZIA) [1]
lateralis, 11

eat

o

“*< in | im
. é wits iM
So hie |

d

int

ars
id
nine

“ee

Ditty’.

P.

Mala HDS VtY ah of
Perit 4
piles

ay yep

ein
a

‘

ae | ages nai

Contributions
of the

American Entomological Institute

Volume 4, Number 2, 1969

MOSQUITO STUDIES (Diptera, Culicidae)

XII. A revision of the Neotropical subgenus Howardina of
Aedes. By O. G. W. Berlin

a NSO Wi qas
AUG 8 14969
C/BRARIED

CONTRIBUTIONS
of the
AMERICAN ENTOMOLOGICAL INSTITUTE

The Contributions are for larger papers on insects. Each paper is a separate number, with
separate pagination and index. Separate numbers aggregating about 500 pages constitute a volume.
Issues appear irregularly, as suitable manuscripts are available.

Copies are sold separately or in subscriptions to complete volumes. Complete volumes are
$12.00. The price of separate numbers varies. Subscribers are billed for each volume with its
beginning number, and receive the parts as issued. Orders for separate numbers that total less
than $8.00 must be accompanied by payment. Address orders or correspondence to the American
Entomological Institute, 5950 Warren Road, Ann Arbor, Michigan, U. S..A. 48105

Parts of Volume 1, with prices

No. 1. Dasch, Clement E., 1964. The Neotropic Diplazontinae (Hymenoptera, Ichneumonidae).
75 pages, 69 figures. Price: $2.25, postpaid.
No. 2. Mosquito studies (Diptera, Culicidae), 1965.
I. Belkin, John N., etal. A project for a systematic study of the mosquitoes of
Middle America. |
Il. Belkin, John N., etal. Methods for the collection, rearing and preservation of
mosquitoes.
78 pages, 4figures. Price: $2.25, postpaid.
No. 2a. Same as no. 2, but in Spanish. Price: $2.25, postpaid.
No. 3. Matthews, Robert W., 1965. The biology of Heriades carinata Cresson. (Hymenoptera,
Megachilidae) 33 pages, 23 figures. Price: $1.00, postpaid.
No. 4. Mosquito studies (Diptera, Culicidae), 1965.
Il. Ramalingam, Shivaji, and John N. Belkin. Two new Aedes from Tonga and Samoa.
IV. Belkin, John N. The mosquitoes of the aibaiue Lah ieee) Museum of Salem
expedition to the southwest Pacific, 1956.
34 pages, 3 figures. Price: $1.00, postpaid.
No. 5. Mosquito studies (Diptera, Culicidae), 1965.
V. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.
Mosquitoes originally described from Middle America. 95 pages. Price: $2.5 postpaid.
No. 6. Mosquito studies (Diptera, Culicidae), 1966.
VI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.
Mosquitoes originally described from North America. 39 pages. Price: $1, 00, postpaid.
No. 7. DeLong, Dwight M. and Paul H. Freytag, 1967. Studies of the world Gyponinae
(Homopotera, Cicadellidae). A synopsis of the genus Ponana.
86 pages, 257 figures. Price: $2.50, postpaid.
No. 8. Contributions to the mosquito fauna of southeast Asia, 1967.
I. Delfinado, Mercedes D. The genus Aedes, subgenus Neomacleaya Theobald in
Thailand. 55 pages, 20figures. Price: $1 1.5, postpaid.

Parts of Volume 2, with prices

No. 1. Contributions to the mosquito fauna of southeast Asia, 1967.
Il. Bram, Ralph A. The genus Culex in Thailand (Diptera: Culicidae).
296 pages, 104 figures. Price: $9.00, postpaid.

No. 2. Schuh, Toby, 1967. The shore bugs (Hemiptera: Saldidae) of the Great Lakes region.
35 pages, 30 figures. Price: $1.25, postpaid.

No. 3. Snider, Richard J., 1967. The chaetotaxy of North American Lepidocyrtus s. str.
(Collembola: Entomobryidae). 28 pages, 57 figures. Price: $1.00, postpaid.

No. 4. Contributions to the mosquito fauna of southeast Asia, 1968.
Il. Delfinado, Mercedes D. The genus Aedes, subgenus Neomacleaya Theobald of
southeast Asia. 76 pages, 26figures. Price: $2.25, postpaid.

No. 5. Contributions to the mosquito fauna of southeast Asia, 1968.
IV. Knight, Kenneth L. Species of the subgroup Chrysolineatus of group D, genus
Aedes, subgenus Finlaya Theobald. 45 pages, 12 figures. Price: $1.50, postpaid.

Parts of Volume 3, with prices

No. 1. Mosquito studies (Diptera, Culicidae), 1968.
VII. Belkin, John N. The Culicidae of New Zealand.
182 pages, 30 figures. Price: $5.50, postpaid.
No. 2. Mosquito studies (Diptera, Culicidae), 1968.
VI. Zavortink, Thomas J. A prodrome of the genus Orthopodomyia.
221 pages, 41 figures. Price: $6.75, postpaid.
No. 3. Fleming, Richard C., 1968. Head musculature of sphinx moths (Lepidoptera:
Sphingidae). 32 pages, 37 figures. Price: $1.00, postpaid.
No. 4. Mosquito studies (Diptera, Culicidae), 1968.
IX. Belkin, John N. The type specimens of New World HORAN OCE in European
museums. 69 pages. Price: $2.00, postpaid.

(Continued on back cover)

MOSQUITO STUDIES (Diptera, Culicidae)

XII. A REVISION OF THE NEOTROPICAL SUBGENUS

HOWARDINA OF AEDES’

By

O.G.W. Berlin?

CONTENTS
INTRODUCTION. . fs
MATERIAL AND METHODS . ) A
GENERAL CONSIDERATIONS . 5
Taxonomic characters . 5
Bionomics 6
Distribution . 7
Systematics and Evolution , g
Affinities . 7
Medical Importance . 8
TAXONOMIC TREATMENT . 8
Subgenus Howardina sheca HA tengo Sh GP eB RR died nik gibt ag ee Partie
re 1 Nh A ae ian el Ur, i er ie Rae he ate bik baat tte ag iggy abe bare
PRY OTR CEO 2 38 nse os), RN Oa a hei wa es eee
Keys to species. . Cede Wai i ET isu omen tae RE ee Cine a ue ae
1. Aedes (H.) fulvithorax (Lutz) .. mae Be easier OAM ee cmon CPN
2. Aedes (H.) septemstriatus Dyar and Knab pa Alias ae RAISE Aa Ree eR
3. Aedes (H.) arborealis Bonne-Wepster and Bonne. . ....... 22
OR ORUTE ) CUPS GOGO DBD oo ee ON re ae a
5 eres. ) 5p. SOS POP oF. He erie ae ie eae

1Contribution from project “Mosquitoes of Middle America” supported by U.S. Public Health
Service Research Grant AI-04379 and U.S. Army Medical Research and Development Command
Research Contract DA-49-193-MD-2478. Based on Ph.D. dissertation submitted to the University
of California, Los Angeles and supported in part by National Science Foundation Grant GB 3871
for field studies.

Department of Zoology, University of California, Los Angeles, California 90024.

Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Walkeri Section

Keys to groups and soecies

Walkeri Group. . .
6. Aedes (H.) walkeri (Theobald)
7. Aedes (H.) argyrites Dyar and Nunez Tovar ;

Aurites Group . . poe eae
8. Aedes (H.) aerihes (Theobald).
9. Aedes (H.) grabhami,n.sp. .

10. Aedes (H.) inaequalis (Grabham)

11. Aedes (H.) stenei Thompson .

Ioliota Group . .

12. Aedes (H.) joliota Dyvar and oe

13. Aedes (H.) cozumelensis Diaz Najera .

Albonotatus Group. . .

14. Aedes (H.) albonotatus (Coquille)

15. Aedes (H.) bahamensis, n.sp.

Busckii Group . .

16. Aedes (H.) busckti (Coquillett)

Sexlineatus Section .

Keys to groups and species
Quadrivittatus Group . .

17. Aedes (H.) quadrivittatus (Coquillett)
18. Aedes (H.) lorraineae, n.sp.
19. Aedes (H.) brevis, n.sp. .

20. Aedes (H.) spinosus,n.sp. .
Sexlineatus Group . .

21. Aedes (H.) sexlineatus (Theobald).
Whitmorei Group. . .
22. Aedes (H.) whitmorei Dunn .

23. Aedes (H.) pseudodominicii Komp.
Allotecnon Group.

24. Aedes (H.) allotecnon Kumm, Komp and Ruiz .

25. Aedes (H.) guerrero, n.sp.

26. Aedes (H.) guatemala, n.sp.
Eleanorae Group . .

27. Aedes (EH) eleanorae, n. sp.

28. Aedes (H.) marinkellei, n.sp. .

29. Aedes (H.) leei, n.sp. .

30. Aedes (H.) osornoi, n.sp.
31. Aedes (H.) ecuadoriensis, n.sp. .
32. Aedes (H.) brevivittatus, n.sp.
Aurivittatus Group . .
33. Aedes (H.) aurivittatus Cerqueira ;
34. Aedes (H.) martinezi,n.sp..

35. Aedes (H.) vanemdeni Martini

REFERENCES CITED
FIGURES
INDEX TO SCIENTIFIC NAMES

Berlin: Subgenus Howardina 3

INTRODUCTION

The exclusively Neotropical subgenus Howardina has been recognized as a distinct

entity in the widespread genus Aedes since Dyar’s classification of the American spe-
cies of the genus (Dyar 1920:103-106) but no attempt has been made to date to de-
velop an internal classification of this subgenus. Nearly all species assigned to How-
ardina have been characterized entirely on the pattern of ornamentation of the
adults. The immature stages were largely unknown. In recent years, considerable as-
sociated reared material has accumulated and it is now possible to attempt a compre-
hensive revision of the subgenus based on all stages.

The taxonomic history of Howardina is brief and relatively simple except for con-
fusion with other generic group taxa (Finlaya, Haemagogus, Stegomyia, etc.) which
need not be detailed here. Howardina was proposed by Theobald (1903:287) as a
distinct genus with 2 included species, walkeri and greenii, and with the following
definition: ‘“‘Head clothed with flat and narrow curved scales, the latter forming a
median area as in Aedes. Scutellum with narrow curved scales; the midlobe with four
posterior border-bristles. Wings with lateral vein-scales large for the size of the wings,
long, rather thin, the median vein-scales small. Palpi of the 2 minute, four-jointed;
the penultimate joint long, longer than the two basal ones; apical joint very minute.
6 unknown. All small species, and with ornamented thorax.” Dyar (1905:44) selected
walkeri as the type species of Howardina. Coquillett (1906:312) recognized the im-
portance of the simple claws, a character mentioned in the original description of
walkeri. By 1910 Theobald recognized 7 species in the genus, described by himself
and by Grabham, but 3 of these were subsequently removed to the genus (later sub-
genus) Finlaya. Howard, Dyar and Knab (1912a:65,70) synonymized Howardina
with Aedes but Dyar (1920:103) elevated it to subgeneric rank within Aedes on the
basis of male genitalic characters (rudimentary claspette). All other contributions be-
tween 1910 and 1932 were merely descriptions of individual species. Edwards (1932:
155-156) in his classic revision of the mosquitoes of the world listed 13 species in the
subgenus Howardina and gave the following brief diagnosis which has been followed
by all authors to the present: “‘most closely related to Finlaya, from which it differs
chiefly in the simple claws of the ¢; less prominent eighth sternite of 9 abdomen; and
less developed claspette of 6 hypopygium, the stem being very short and the append-
age bristle-like. Vertex in all species with a median stripe of narrow scales reaching
forward to eyes; scutellar scales narrow. Thorax conspicuously ornamented with
white or golden lines. Larva.- Similar to that of Finlaya; a more or less distinct small
chitinous plate present on each side of the ventral brush.’ Subsequent to Edwards’
revision more confusion than progress has been accomplished, particularly in errone-
ous identifications of species. In the world catalog of mosquitoes (Stone, Knight and
Starcke 1959:175-176) 16 species were recognized on the basis of published records.

Included in Edwards’ (1932:155-156) list of species of Howardina were dominicii
and whitmorei for which Komp (1936:71) subsequently erected the subgenus So-
peria, with dominicii as the type species. Unfortunately Komp misidentified domi-
nicii which in reality is conspecific with sexlineatus in the subgenus Howardina.
Komp based the new subgenus primarily on the complete absence of a claspette in
the male genitalia. Soperia has been considered as a distinct subgenus to the present
and contained 3 species in the world catalog (Stone, Knight and Starcke 1959:173),
one of which, tracei, was synonymized with busckii in the subgenus Howardina by
Belkin, Schick and Heinemann (1965:33).

4 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

In the present study only the subgenus Howardina is recognized; the species for-
merly placed in Soperia of authors form a distinct group included in the Sexlineatus ©
Section, 1 of 3 primary divisions of Howardina. A total of 35 species is now recog-
nized in the subgenus: 5 species in 1 group in the Fulvithorax Section; 11 species in
5 groups in the Walkeri Section; and 19 species in 6 groups in the Sexlineatus Sec-
tion.

I wish to express my gratitude to Dr. John N. Belkin for suggesting this project
and directing the study. I am indebted to Dr. Alan Stone of the U.S. National Muse-
um for the loan of material and for information on type specimens; to Dr. P.F. Mat-
tingly for the loan from the British Museum; and to Dr. A. Martinez, Dr. A. Spielman
and Dr. A. Diaz Najera for specimens from Argentina, Bahamas and Mexico respec-
tively. I also thank Pedro Galindo, Gorgas Memorial Laboratory, Panama, and Dr. I.
Goodbody, Department of Zoology, University of West Indies, Kingston, Jamaica,
for the courtesies extended to me during field trips to these countries. Finally I wish
to thank Dr. R.X. Schick, Dr. S. Sirivanakarn, Dr. T.J. Zavortink and Miss Sandra
Heinemann for the valuable suggestions and assistance and to Miss Sheila Bernstein
for the preparation of the final copy for reproduction.

MATERIAL AND METHODS

MATERIAL. This study is based largely on extensive collections of adults and as-
sociated immature stages accumulated at the University of California, Los Angeles
for the project ‘““Mosquitoes of Middle America’’ (Belkin, Schick et al, 1965) and also
on material from the U.S. National Museum and the British Museum (Nat. Hist.)
where most of the types of previously described species are deposited. Additional
critical material was obtained specifically for this study in Jamaica in November and
December 1966 following the methods of collection and rearing outlined by Belkin,
Hogue et al (1965).

For this revision a total of 10,949 specimens were studied, including 1,240 6,
(3,527 2, 2,412 pupae, 3,770 larvae, with 1,796 individual rearings (810 pupal, 645
larval, 341 incomplete).

TAXONOMIC PROCEDURE. The taxonomic methods used in this study are the
Classical phylogenetic ones, simple in principle but difficult in practice. Following the
recognition of species on the basis of a series of constant correlated characters in all
stages, the different species were compared minutely and those showing the greatest ©
similarities in all stages were interpreted as members of primary phyletic lines or spe-
cies groups, called simply Groups here. These groups were compared in turn to estab-
lish the more inclusive Sections within the subgenus.

DESCRIPTIONS. The method of presentation, terminology and abbreviations
used in the descriptions of taxa in general follow Belkin (1962). A few special terms
used for the pattern of ornamentation of adults are explained in the section on taxo-
nomic characters in the following chapter. ;

ILLUSTRATIONS. All available stages of every species are figured, the larvae and
pupae in full and the adults in pertinent details only. Each illustration is based on a
topotypic specimen, whenever available, but the character states shown represent the
modal condition determined usually from at least 10 topotypic specimens. In the il-
lustrations of the ornamentation of mesonotum and pleuron, dotted lines represent
golden or creamy scales and dashed lines denote white or silvery scales. |

Berlin: Subgenus Howardina 5

DISTRIBUTION DATA. In the distribution lists, the arrangement of countries is
from north to south and west to east and the localities in each country are listed al-
phabetically under the major political subdivisions. Included in these lists are all the
records of material | have examined as well as published records from the literature
which I consider reliable. In these lists, P and L denote whole pupa and larva and p
and |, pupal and larval skins respectively.

DEPOSITORIES. The abbreviations used by Stone, Knight and Starcke (1959)
for institutions in which types are deposited are followed here to indicate the loca-
tion of the material studied. To these are added the Institute of Jamaica [IJ] and the
Department of Zoology, Los Angeles, California [UCLA]. The bulk of the material
will remain at UCLA. The types of all new species described here will be deposited
in the U.S. National Museum.

GENERAL CONSIDERATIONS

TAXONOMIC CHARACTERS. The following discussion is limited to external
morphological characters. Other important taxonomic characters, ecological and geo-
graphical, are discussed below in the sections on Bionomics and Distribution respec-
tively. : hae,

In the adults, the most important taxonomic characters pertain to the pattern of
ornamentation of the mesonotum. As indicated in the diagnosis of the subgenus, the
_ pattern usually consists of lines of light scales superimposed’ on a background of dark
scales. Each section, each group and most species have a distinctive pattern of orna-:
mentation. It was necessary to develop a special terminology for these light lines as
no suitable one was available. This terminology is illustrated in figs.8 and 9 and ex-
plained below. The acrostichal line is the single median longitudinal mesonotal line
and is developed only in some members of the Fulvithorax Section; it ends in front
of the prescutellar space and is more or less distinctly continued caudad as a pair
of lateral prescutellar lines. The median prescutellar line is developed only in the
Walkeri Section and is in the median longitudinal position in the “‘bare’’ prescutel-
lar space. The inner dorsocentral lines are located mesad of the dorsocentral row of
bristles in a submedian longitudinal position on each side; these lines reach the scu-
tellum only in the Sexlineatus Section; the posterior parts (prescutellar) of these
lines more or less correspond to the lateral prescutellar lines found in some members
of the Fulvithorax Section. The outer dorsocentral lines are located laterad of the
dorsocentral row of bristles in a sublateral longitudinal: position; they are repre-
sented to a varied degree in all sections and usually reach the lateral lobes of the scu-:
tellum in the Walkeri and Sexlineatus Sections. The lateral marginal lines border
the mesonotum on each side to a varied degree from the anterior promontory
to the wing root; they may be broken into one or more prescutal and antealar seg-
ments. Transverse diagonal posterior fossal lines may connect the lateral marginal to
the outer dorsocentral on each side. Longitudinal supraalar lines may also be devel-
oped and connected to the lateral marginal lines on each side caudad of the scutal an-
gle. In fulvithorax and to a lesser extent in arborealis of the Fulvithorax Section,
the ornamentation is primarily light and may be considered to have been developed
by the expansion and fusion of the light lines which are discrete in the other sec-
tions, particularly in the Walkeri Section.

Other significant taxonomic features of adult ornamentation are: (1) on the head,
scaling of the vertex and of the torus and first flagellar segment of the antenna; (2)

6 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

scaling of the paratergite; (3) distribution of the scales on the pleuron, particularly
ppn, stp and mep; (4) on the legs, light scaling of the foretibia, midfemur and hind-
femur, and the distribution and extent of light scales on the tarsal segments.

The comparatively simple male genitalia of Howardina snow relatively few obvious
taxonomic characters; even specific differences are usually subtle and sometimes not
apparent. The reduced claspette is more or less characteristically developed in each
section and is absent in the Whitmorei Group. Other significant taxonomic characters
appear to be: (1) extent of separation of the base of the sidepieces midsternally, (2)
width and length of the aedeagus in relation to the proctiger and (3) the shape of the
aedeagus in dorsal aspect. Only 3 species (bahamensis, whitmorei and pseudodomi-
nicii) have genitalia strikingly differentiated from the others in features of the side-
pieces.

The pupae are quite similar in the subgenus but a few characters separating the 3
sections have been noted in: (1) development of cephalothoracic hair 5-C, (2) posi-
tion of abdominal hair 2-[V-VI and (3) development of abdominal hair 9-VI. On the
other hand, reliable group characters within a section have not been found and spe-
cific differences are rather subtle and not always reliable.

The 4th-instar larvae show many taxonomic characters at the section and group
levels usually concordant with those of adult ornamentation, and many reliable spe-
cific differences. Among the important taxonomic characters are: (1) presence or ab-
sence of marginal spicules on the anal saddle, (2) comb scales in a single row or in a
patch, (3) ventral brush with 5 or 6 pairs of hairs, (4) prothoracic hair 13-P present
or absent, (5) presence or absence of integumentary spicules on the thorax and/or
abdomen, and (6) numerous features in the development of hairs, particularly on the
head.

BIONOMICS. All members of the subgenus, as far as known, are container breed-
ers. Immature stages have been reported from rockholes, treeholes, broken bamboo,
plant parts on the ground, leaf axils of bromeliads and aroids, and flower bracts of
heliconias. A few species invade artificial containers of various types. Only the
Walkeri Section utilizes the full range of breeding habitats; the Fulvithorax Section
is apparently restricted to treeholes and broken bamboo and the Sexlineatus Sec-
tion is known to breed only in bromeliad axils. :

In common with other ornamented mosquitoes, species of Howardina are appar-
ently largely diurnal in activity (Bates 1944:159-170; 1949:16,61; Carpenter, Galin-
do and Trapido 1952:163; Trapido, Galindo and Carpenter 1955:530,537,538; For-
attini 1965:401-403). To supplement the published information, I made a full 24-
hour observation on the biting activity of females near Porus, Manchester, Jamaica
at an elevation of approximately 1000 ft, where 4 species of Howardina were pres-
ent. Biting-landing collections were made at hourly intervals at ground level under
bromeliad laden trees. As seen in fig.7 there appear to be specific or group differ-
ences in activity since walkeri shows 2 definite peaks, a higher one in the morning,
and members of the Aurites Group show only a moderate afternoon peak.

Although a number of species are known to bite man in forested areas, the definite
altitudinal stratification of different species suggests that their natural hosts are a
variety of vertebrates characteristic of a particular stratum in the forest. Members of
the Quadrivittatus Group have been collected mainly in the forest canopy (Trapido
and Galindo 1957:122,123,124,125,131), while members of the Fulvithorax Group
are predominantly forest floor biters (Trapido and Galindo 1957:131); members of
the Sexlineatus Group (Trapido and Galindo 1957:132) and Whitmorei Group (Bates
1944:166), on the other hand, demonstrate inconsistent behavior.

Berlin: Subgenus Howardina 7

DISTRIBUTION (fig.1). Howardina is strictly a Neotropical taxon, centered in
the American Mediterranean region in the sense of Belkin (1962:562). At present its
known northern limit is Grand Bahama Island in the east and the Sierra Madre del
Sur and the southern end of the Sierra Madre Oriental in Mexico in the west. The
southern limit appears to be in northern Argentina along the Andes; the subgenus ap-
parently does not reach the mouth of the La Plata river in the east.

The distributions of the 3 sections are largely complementary. The Walkeri Sec-
tion is almost entirely confined to the islands of the Caribbean area with only a
slight overlap with the other 2 in northern Venezuela and Trinidad. The Fulvithorax
Section is found primarily in eastern South America but with a significant overlap
with the Sexlineatus Section in southern Central America. The Sexlineatus Section
is primarily associated with the mountain regions of Central America and western
South America.

SYSTEMATICS AND EVOLUTION. The 35 species recognized in the present
study fall into 12 largely well-defined, discrete groups of 1 or more species each. On
the basis of morphological similarity the 12 groups form 3 very distinct sections in
the subgenus, the Fulvithorax Section with only 1 group (Fulvithorax) of 5 species
(1 not named), the Walkeri Section with 5 groups (Walkeri, Aurites, Ioliota, Albo-
notatus, and Busckii) of 11 species, and the Sexlineatus Section with 6 groups
(Quadrivittatus, Sexlineatus, Whitmorei, Allotecnon, Eleanorae and Aurivittatus) of
19 species.

As indicated in the preceding discussions, as well as in the taxonomic treatment of
each section, there is practically no overlap in diagnostic morphological characters
among the 3 sections, little overlap in distribution, and strong differentiation in the
breeding sites between Fulvithorax Section and the Sexlineatus Section. It appears,
therefore, that the 3 sections represent 3 old phyletic lines.

In spite of this clear-cut divergence, there is a remarkable basic similarity in all
stages in all 3 sections. This is undoubtedly indicative of true affinity and some kind
of monophyletic origin. Two general explanations for the development of 3 major
phyletic lines in Howardina are possible. First, would be serial or independent devel-
opment of 2 of the sections from the third; this is not plausible because each of the
lines has retained primitive features absent in 1 or 2 of the others. Second, would be
a tripartite splitting of one ancestral species. This interpretation is supported by the
present distributions of the 3 sections and I therefore suggest on this basis that the
original stock was divided into 3 segments: (1) one in the Caribbean, giving rise to
the Walkeri Section, (2) one in Central America, giving rise to the Sexlineatus Sec-
tion, and (3) one in northeastern South America, giving rise to the Fulvithorax Sec-
tion.

Subsequent speciation and evolution have followed different patterns in each sec-
tion, probably because of different environmental conditions. In the Fulvithorax
Section, which occupies primarily eastern South America where relatively few bar-
riers to dispersal have existed, at least in relatively recent geological times, there
has been little differentiation and speciation. The Walkeri Section shows typical
complex insular speciation of a variety of divergent types, including relict forms.
In the rather homogeneous Sexlineatus Section the extensive speciation reflects the
barriers and environmental diversity associated with the complex orogenies of the
cordilleras of western America.

AFFINITIES. The general similarity between the subgenera Howardina and Fin-
laya has been recognized by many culicidologists, notably Edwards (1932:156), and
as indicated in the introduction, some species of Finlaya were originally included in
Howardina, primarily on the basis of the ornamentation of the adults. The chief dif-

8 Contrib. Ate, But Inch aca danas 2, 1969

ferences from Finlaya are the untoothed claws and less prominent sternite VIII in
the females and the claspette poorly developed or completely absent in the males. In
the last feature, Howardina shows little difference from the Oriental dissimilis sub-
group of Finlaya (Knight and Marks 1952:516-517). The immature stages of How-
ardina also bear a general resemblance to those of Finlaya and are very similar to
those of some species of the genus Haemagogus, which is also restricted to the Neo-
tropical region and may have evolved from a Finlaya stock “‘through some such spe-
cies as A. (F.) leucocelaenus”’ (Edwards 1932:178). The following unusual combina-
tion of features in the larvae of Howardina may be diagnostic for the subgenus: ab-
dominal hair 12-I never developed, hair 9-I always ventrad of hair 7-I, and tubercles
of prothoracic hairs 5-7-P usually distinctly joined (except in Fulvithorax Section).

Little can be done now to clarify the affinities of Howardina chiefly because the
immature stages of the majority of annectent groups in the tribe Aedini are very
poorly known both in the Old World and the New World. In spite of some apparent
slight overlap with the subgenus Finlaya and the genus Haemagogus, Howardina is a
very distinct group, and should be retained as a subgenus of Aedes.

MEDICAL IMPORTANCE. Little information is available on the medical impor-
tance of species of Howardina. A few of the more abundant species were investigated
in connection with studies on the vectors of jungle yellow fever. Experimental stud-
ies with fulvithorax showed that while it could be infected with and maintain the
virus, it was not able to transmit it naturally through bites (Davis and Shannon 1931;
Whitman and Antunes 1937). Similar results were obtained with other species of
Howardina by Boshell-Manrique and Osorno-Mesa (1944), Bugher, Boshell-Manrique
et al (1944) and Aitken (1960). Bates (1949:205) has suggested that the incubation
period of this virus in some mosquitoes may be a long one and it is possible that
Howardina species may transmit the yellow fever virus if the incubation period is
prolonged at favorable temperatures. Galindo (1964) reported the isolation of an
unidentified virus from guadrivittatus in Panama. Since some species of Howardina,
especially members of the Walkeri Section (Walkeri, Aurites and Busckii Groups),
attack man readily and are very abundant, a more thorough investigation of the
role of these species as vectors of arboviruses should be made.

TAXONOMIC TREATMENT

Subgenus HOWARDINA Theobald

1903.Howardina Theobald, 1903:287. TYPE SPECIES: Culex walkeri Theobald, 1901, Jamaica;
the second of 2 included species, selection of Dyar (1905a:49). 7

1936.Soperia Komp, 1936b:71. TYPE SPECIES: Aedes dominicii Rangel and Romero-Sierra,
1907, Venezuela, original designation. NEW SYNONYMY.

Aedes (Howardina) of Dyar (1920:10); Bonne and Bonne-Wepster (1925 :366); Dyar (1928:103);
Shannon (1931:125); Edwards (1932:155); Gerry (1932:40); Lane (1939:98); Anduze (1941:
14); Bates (1949:320); Lane (1949:255,261); Del Ponte, Castro and Garcia (1951:231,239);
Lane (1953:706); Levi-Castillo (1953:40); Horsfall (1955:471); Martinez and Prosen (1955:21-
22); Peyton, Galindo and Blanton (1958:95-100); Stone, Knight and Starcke (1959:173-174);
Martinez, Prosen and Carcavallo (1960:25); Belkin (1962:340); Stone (1963:127); Prosen, Car-
cavallo and Martinez (1964:103); Forattini (1965:397-401); Cova Garcia, Sutil and Rausseo
(1966a:in part; 1966b:333); Porter (1967:38); Stone (1967:208).

Berlin: Subgenus Howardina 9

Aedes (Soperia) of Lane (1939:98); Anduze (1941:14); Senevet and Quievreux (1941:258); Andu-
ze (1943:193; 1947:357); Lane (1953:700); Levi-Castillo (1953:40); Komp (1956:38); Stone,
Knight and Starcke (1959:173); Forattini (1965 :404,460,462); Cova Garcia, Sutil and Rausseo
(1966b:338-339). ,

Howardina of Blanchard (1905:399,415); Dyar (1905:49); Theobald (1905:21); Theobald and
Grabham (1905:7), Coquillett (1906b:16,26); Dyar and Knab (1906b:188); Theobald (1907:
21,149,214); Coquillett (1910:553); Theobald (1910:220); Surcouf and Gonzalez-Rincones
(1911:102,139-141; 1912:276).

Aedes in part of Howard, Dyar and Knab (1912:65,70; 1917:613); Dyar (1918:71).

FEMALES (fig.16). Small to medium sized ornate species; usually blackish to dark
brown with white, silver, creamy, yellowish or golden markings on head, thorax and
abdomen and with white basal tarsal markings. Head: Eyes distinctly separated be-
tween antennae. Decumbent scales predominantly broad and flat; narrow scales usu-
ally largely restricted to median longitudinal light stripe of varied width running from
lower end of interocular space to occiput; lower end of interocular space always with
a tuft of very long outstanding narrow bristlelike striated scales; orbital line of nar-
row light scales, varied in width and length, frequently developed. Erect scales short,
usually forked apically, restricted to occiput or extending on to the disc and sides of
vertex. One pair (rarely 2) of interorbital bristles always differentiated; a single row
of orbitals, upper usually 4(3-6) pairs, heavier, longer and more closely spaced than
lower. Clypeus bare, deep brown to black. Proboscis slender; subequal to or distinct-
ly longer than forefemur; entirely dark scaled; with a few basal bristles. Palpus al-
ways short, about 0.15 to 0.2 of proboscis; 5-segmented, segments 1 and 2 anky-
lozed, segment 5 minute; segments 1 and 5 without scales, remainder entirely or pre-
dominantly dark scaled, light scales usually restricted to apex of segment 4. Antenna
distinctly shorter than or subequal to proboscis; torus dark brown to black, with
light to dark scales always present, sometimes indistinct, usually restricted to mesal
surface; flagellar segment 1 only slightly longer than segment 2, always with a few
small scales on mesal surface; flagellar segments 2-13 usually with 6 moderate bris-
tles in basal whorls. Thorax: Integument dark brown to almost black. Mesonotum
usually with a background of short, narrow curved scales auburn to black in color
and with a varied pattern of lines of light scales (except in fulvithorax and arborealis
in which light scales predominate) as follows: (1) median acrostichal, (2) inner dorso-
central, (3) outer dorsocentral, (4) lateral marginal, (5) prescutal, (6) posterior fossal,
(7) supraalar, (8) median prescutellar and (9) lateral prescutellar. Acrostichal bristles
usually not developed on disc (present in busckii); dorsocentrals, prescutellars and
supraalars always present, variously developed; 1 posterior fossal and 1 parascutellar
always developed. Median scutellar lobe with 4,5 large marginal bristles and with dark
broad flat scales laterally and narrow light scales in varied median longitudinal line;
lateral lobe with 3-5 large marginal bristles and usually with narrow scales only. Para-
tergite bare or with scales. Bristles present on apn, ppn, ppl, psp, stp, pra and upper
mep; lower mep bristles always absent; lowermost stp bristles distinctly ventrad of
lower margin of mep. Pleural scaling dingy white, white or silvery; always present on
apn, ppl, base of pra, stp and mep; ppn, psp, and ssp with or without scales. Legs:
Coxae with light scales on external surface; trochanters with mesoventral light scales. .
White or silvery knee spots prominent on midleg and hindleg, slightly indicated on
foreleg. Midfemur and hindfemur with or without distal light spot on anterior surface.
Tibiae completely or predominantly dark; light scaling either as narrow lines or as
patches. Tarsal white rings always basal, usually restricted to proximal segments.
Claws simple on all legs but with minute spicules at base of external surface. Wing:

10 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Veins usually entirely dark scaled, white scales rarely present on remigium; plume
scales restricted to veins Rg, R243, Ry, R3 and M. Abdomen: Tergite I with numer-
ous scales; laterotergite with large white or silvery scale patch. Tergites II-VII with
broad basolateral silvery patches, progressively more removed from base and extend-
ing more tergally on distal segments, on VIII usually merging with its mate middor-
sally; basomedian tergal light scaling present or absent on II-VII. Light scaling of
sternites varied.

FEMALE GENITALIA (fig.16). Deeply retracted into segment VIII, even cerci
usually invisible. Segment VIII partially retracted into segment VII, apex visible, nu-
merous scales and bristles present; tergite about 0.6 length of tergite VII; sternite
about 1.4 of tergite length and about 0.8 of sternite VII. Tergite [X strongly devel-
oped, about 0.5 of tergite VIII, undivided, without bristles. Sigma joined to cowl,
narrow; continuous with poorly sclerotized insula which bears 2 pairs of setae. Cowl
narrow, joined to base of postgenital plate; atrial plates absent. Postgenital plate
shorter than cercus, bilobed. Tergite X not developed. Cercus broad, subequal in
length to tergite [X or a little longer. Spermathecae 3, 1 a little larger than others.

MALES (fig.16). Coloration usually similar to females, sometimes light scales of
torus absent (allotecnon) and rarely proboscis with a ventral light patch (whitmorei);
sexual dimorphism of head appendages marked. Proboscis usually longer than in fe-
males. Palpus porrect, slender, subequal to proboscis; 5-segmented, segments 4 and 5
subequal, narrowed; entirely or predominantly dark scaled, light scales restricted to
lower surface of segments 4 and 5; a few short inconspicuous setae on body of seg-
ment 4 and apex of segments 3 and 5. Whorls of flagellar segments 1-12 very strong-
ly developed, usually with about 40 or more long bristles; flagellar segment 12 and
13 elongate, 12 about 1.25 of 13; torus greatly swollen. Claws of foreleg and midleg
enlarged, unequal; larger claw with a blunt submedian tooth; both claws with or
without an acute basal external tooth and with external basal spicules more numer-
ous than in females, extending on shaft. Hindclaws as in the females.

MALE GENITALIA. Segment VIII long and wide, unspecialized. Segment [IX
moderately to strongly developed; tergite lobes indistinct or prominent and bearing
a few to many moderate to strong setae, connected middorsally by a narrow sclero-
tized bridge, sternite strongly developed, usually with distinct median sclerotization
bearing several pairs of setae distally. Sidepiece well developed; more or less conical
or subcylindrical; mesal membrane developed from base to apex; tergomesal and/or
sternomesal margins rarely (Whitmorei Group) with slight median lobes; with numer-
ous long bristles and scales laterally and ventrally, tergal surface varied. Claspette
rarely completely absent (Whitmorei Group), usually very poorly developed as a
small sternal basomesal lobe bearing 1 or rarely 2 thickened apical setae and some-
times a few other less differentiated setae. Clasper simple, slender; 1 apical spiniform
always present. Aedeagus without teeth; varied in shape, usually large; lateral sclero-
tization sometimes produced as a distal horn or lobe. Proctiger strongly developed,
prominent; paraproct with a simple, large, heavily sclerotized, curved apical tooth;
cercal setae distinct, short.

PUPAE. Cephalothorax: Middorsal ridge moderate; all hairs present, variously de-
veloped; hair 6-C always smaller than and cephalad of 7-C; hairs 8,9-C always widely
separated and far caudad of trumpet base. Trumpet: Not placed on tubercle; varied
in length; index from about 3.0 to 9.0; always without tracheoid; pinna short to long.
Metanotum: Hairs 10,11-C removed from 12-C, moderately close together. Abdo-
men: Hairs 8,10,11-II apparently usually absent; hair 5-II,II] usually caudad of hair
4: hair 1 usually most mesal hair on all segments; hair 9-VIII ventral in position, near

Berlin: Subgenus Howardina 11

caudolateral angle or removed considerable distance cephalad; hairs 1-[X,X apparent-
ly not developed. Tergite IX strongly developed, its posterior margin smoothly
rounded; cercal lobe of female projecting only a short distance beyond genital lobe.
Paddle: Variously developed; midrib usually strongly differentiated; marginal spic-
ules varied.

LARVAE. Head: Width of head capsule subequal to length or slightly greater; la-
brum indistinctly differentiated, very narrow; mouthbrushes very numerous, filamen-
tous; collar well developed, narrow; ventral part of head capsule (labial plate) rather
short; posterior tentorial pit near caudal border; maxillary suture complete, some-
times extending caudolaterad of pit; cephalic border of labial plate truncate; aulaeum
with filamentous spicules; mental plate well developed. Hair O-C distinct, far re-
moved laterad of 1-C; 2-C apparently always absent; 3-C moderately developed; 11,
12,14-C sometimes stellate. Antenna: Short to moderate in length; shaft with or
without short spicules. Hair 1-A single to 3-branched. Thorax: Integument glabrous
or with more or less distinct spicules. Hairs 0,1,3,4,8,13,14-P, 1,13,14-M, 1,3,4,5,8,
13-T sometimes stellate. Hairs 1-3-P on common tubercle, placed close to middorsal
line; tubercles of 5-7-P usually distinctly joined (except in Fulvithorax Section); 9-
12-P far ventrad, poorly developed, none of the hairs distinctly stellate. Hair 5-T
always cephalad of hair 3. Abdomen: Integument as on thorax. Hair 12-I never
developed, 9-I always ventrad of 7-I; hairs 1,2,5,7,9,13 frequently stellate on seg-
ments I-VI; hair 6-I-VI always well developed, long; hair 7-I long. Segment VIIT:
Comb scales in a single regular row to an irregular patch of 3-5 rows. Siphon: Index
varied from 2.0-9.0. Integument smooth, imbricate or spiculate. Pecten in a straight
line, usually at least reaching to middle of siphon. Hair 1-S usually near middle; ac-
cessory hairs not developed. Acus varied or rarely absent (walkeri). Anal Segment:
Saddle always incomplete; integument smooth, imbricate or spiculate; caudal margin
with or without spines; acus absent. Ventral brush with 5 or 6 pairs of hairs, on a
more or less distinctly sclerotized boss, with or without distinct grid bars.

DISCUSSION. For a general discussion of taxonomic characters, bionomics, dis-
tribution, systematics and affinities see the preceding chapter entitled “General Con-
siderations.”

The subgenus Soperia is here synonymized with Howardina. The taxon involved
represents the Whitmorei Group of 2 species in the Sexlineatus Section, character-
ized primarily by the total absence of the claspette in the male genitalia. In oth-
er respects the Whitmorei Group obviously belongs in this section of Howardina and
it is evident that there is a tendency within this subgenus to reduction in the clasp-
ette. Further, the name Soperia cannot be applied to the Whitmorei Group without
recourse to the International Commission on Zoological Nomenclature since the
nominal type species, dominicii, designated by Komp is a synonym of sexlineatus
which obviously belongs to a different taxonomic group. In view of the above-men-
tioned taxonomic interpretation, there is no need at this time to request the Com-
mission to vary the type species of Soperia, and this subgenus becomes a subjective
synonym of Howardina by strict application of the rules in the current code.

KEYS TO SECTIONS

ADULTS

& Prescutellar space with single median golden, yellow or silvery line; parater-
gite without scales ob ee ee WAL ARERI SECTION

12

2(1).

2(1).

3(2).

4(3).

21 y:,

IL),

Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Prescutellar space with paired lateral hie yellow, golden or silvery lines;
paratereideowmitivscdies ig a ee ee eee ee ee

Mesonotum with acrostichal line always dark . . SEXLINEATUS SECTION
Mesonotum with at least acrostichal line yellow or golden, sometimes anteri-
or half or more of mesonotum entirely or largely yellow or golden-scaled

FULVITHORAX SECTION

MALE GENITALIA

Tergal surface of sidepiece with a very dense vestiture of large bristles ob-
scuring claspette area from dorsal aspect (15. bahamensis) . .

. in part WALKERI SECTION

Tergal surface of sidepiece with sparse ‘vestiture of short or moderate bristles

MOL Goscuring claspeite area trom dorsal ASPeCt an... ek ce, ee ee

Claspette completely absent Eee Group) |. e
. in part SEXLINEATUS SECT ION
Claspette represented by1 more or less strongly developed sternomesal lobe or
plaque bearing usually 1 thickened apical bristle, rarely 2 ormore . . .3

Sidepieces separated from each other at base sternally by a distance less than

maximum width of aedeagus . ... . . .SEXLINEATUS SECTION
Sidepieces separated from each other at base Noe by a distance greater
LAall WiAxteI Wicd IY OL. acccaeus 2... ge eae cate ow do cone ee

Claspettes connected across midline by more or less distinct sclerotized bar
Ae Sere Oe eer ae eer FULVITHORAX SECTION
Claspettes not connected across midline . . . . . . WALKERI SECTION

PUPAE

Hair 5-C longer than distance between its base and that of trumpet, always
double; hair 2-[V-VI usually within basal 0.66 of itstergite. . .

. SEXLINEATUS SECTION

Hair 5-C shorter than distance bétween its base and that of trumpet, varied in

branching; hair 2-IV-VI usually distad of basal 0.75 ofits tergite. . . .2
Hair 9-VI minute, less than 0.5 of hair 2-VI. . FULVITHORAX SECTION
Hair 9-VI subequal to or longer than hair 2-VI. . . . WALKERI SECTION

LARVAE

Caudal border of saddle without distinct marginal spicules; comb scales al-
ways in a patch of 3 or 4irregularrows . . FULVITHORAX SECTION
Caudal border of saddle with conspicuous long marginal spicules; comb scales
usually in a single even row, rarely in a patch of 2-4 irregular rows (J6.
|

Hair 13-P notdeveloped ........... .. . .WALKERI SECTION
Oe eee er ee OE ete ace hd cia a oe do sane beech: Ga

Berlin: Subgenus Howardina 13

3(2). Hair 9-P poorly developed, single (Ioliota Group). .
coo Ieper WALKERI SECTION
Hair 9-P strongly developed, at Jeast A-branched . SEXLINEATUS SECTION

FULVITHORAX SECTION

FEMALES. Medium sized species, wing 2.8-3.6 mm. Head: Scales predominantly
or entirely light, submedian and lateral dark patches sometimes large. Erect scales
present on disc of vertex. Torus yellowish to dark brown, with a few moderately
broad dark scales mesally. Palpus entirely dark scaled. Thorax: Mesonotum varying
from nearly entirely golden or yellow to predominantly dark bronzy but with at.
least golden acrostichal line developed; median prescutellar light line not developed.
Paratergite with scales. Acrostichal bristles absent on disc. Pleural scales silvery or
dingy golden; ppn, psp, and ssp with scales; most of mep covered with large continu-
ous scale patch. Legs: Midfemur and hindfemur without distinct distal light spot on
anterior surface. Apex of tibia and base of tarsal segment 1 of all legs with white
scales; all other tarsal segments dark. Wing: Veins entirely dark scaled. Abdomen:
Sternites II-[V without dark scales.

MALES. Larger claw of foreleg and midleg with submedian tooth only.

MALE GENITALIA. Lobes of tergite [IX indistinct to moderately prominent, with
numerous strong setae. Sidepieces separated from each other at base by a distance
greater than diameter of aedeagus at that level; tergomesal and sternomesal margins
without any indication of lobes. Claspette prominent, connected at base with its
mate by more or less distinct sclerotized bar; apical and subapical bristles varied in
development.

PUPAE. Cephalothorax: Hair 5-C distinctly shorter than distance between its base
and that of trumpet. Abdomen: Hair 5-IV,V usually single, extending beyond apex
of tergite following; 2-IV-VI usually distad of 0.75 of its tergite; 9-II-VI cephalad of
hair 6; 9-VII moderate to strong, close to hair 6, not removed cephalad.

LARVAE. Stellate hairs moderately developed. Antenna: Short; without spicules;
hair 1-A single. Thorax: Integument glabrous; hairs 5-7-P on separate tubercles; hair
9-P usually single or double; hair 13-P absent. Abdomen: Integument glabrous; hairs
2,5,9-II-VII usually stellate, with few branches; hair 8-VII cephalodorsad of 9-VII.
Segment VIII: Hair 5 stellate. Comb scales always in a patch of 3,4 irregular rows;
individual scales rounded apically. Siphon: Index 2.2-2.3. Integument glabrous; pig-
mentation very strong. Acus distinct. Pecten teeth with basal denticles. Anal Segment:
Saddle without marginal spicules. Hair 1-X short to moderate, double or triple; hair
3-X single, 2-X multiple Ventral brush with 6 pairs of hairs arising from a strongly
sclerotized boss.

DISCUSSION. The Fulvithorax Section is probably the most generalized lineage
in Howardina and is strongly differentiated in all stages from all the other sections of
the subgenus. The pattern of ‘ornamentation of the mesonotum of the adults is
unique in the subgenus and shows considerable variation whereas it is stabilized in the
other sections. The male genitalia are readily recognized by the widely spaced side-
pieces and the relatively large prominent claspette. The larvae are immediately sepa-
rated from all the other sections by the characteristic comb patch and the absence of
caudal marginal spicules on the anal saddle. The pupal stage has the simplest chaeto-
taxy in the limited branching and the relatively small size of the more conspicuous
hairs. In nearly all these characters, the Fulvithorax Section shows the closest simi-

14 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

larity to the subgenus Finlaya from which Howardina was probably derived. Further-
more, as in typical American Finlaya, the natural breeding sites of the Fulvithorax
Section are treeholes and broken bamboo.

The Fulvithorax Section presents a challenging taxonomic problem, which, unfor-
tunately, I have not been able to resolve satisfactorily because of the absence of
reared associated material from nearly all the critical areas of its wide reported distri-
bution (fig.2) which is largely confined to the Atlantic drainages and extends from
Nicaragua to southern Brazil. Formerly, 3 species with nearly overlapping distribu-
tions were recognized almost entirely on the basis of the extent and arrangement of
the golden scaling of the mesonotum. The name fulvithorax was applied to forms
with the mesonotum nearly entirely golden, septemstriatus to specimens with dis-
crete lines of golden scales separated by dark scales, and arborealis to specimens in-
termediate in this respect. This situation suggested that only 1 very variable species
was involved. However, closer analysis of the adults available for the present study
and the examination of associated reared immature stages from 3 areas reveals a
much more complex situation.

In the present study, I am recognizing only 1 group with 5 different one in the
Fulvithorax Section: fulvithorax, with a wide distribution from Venezuela and Trini-
dad to southern Brazil and occurring a considerable distance inland in Brazil and
Peru; septemstriatus, restricted to lowland and moderate elevations in Nicaragua,
Panama and Colombia; arborealis, from the Guianas and the lower Amazon; aureo-
lineatus, primarily from the state of Bahia but apparently extending north to Para
and south to Rio de Janeiro; and the unnamed Iquitos form from Peru which may be
an inland type of arborealis. A. fulvithorax, septemstriatus and arborealis all show
more or less distinct north-south clines in the extent of mesonotal golden scales, the
southern populations being in general darker. In the case of fulvithorax, the inland
populations also tend to be darker than the coastal. In any given area, only 2 of these
forms appear to be present, except that 3 may occur in Para, and in all instances the
different sympatric forms are quite distinct, without any intermediates. This suggests
that distinct species are involved. However, the material available for this study is
very limited and it is possible that the 5 forms recognized here are only subspecifical-
ly distinct and that there might be intergradations between them. It is even possible
that only 1 very variable species is involved. The problem cannot be definitely re-
solved without progeny and individual rearings from areas of contact and overlap.

KEYS TO SPECIES

ADULTS

]. Mesonotum usually entirely golden-scaled from anterior promontory to wing
root, rarely a few darker scales along dorsocentral lines posteriorly; no
corispicuous dark bronzy scalesonfossa. ... . . .. . JL. fulvithorax

Mesonotum not entirely golden-scaled in anterior portion, with at least some
conspicuous dark bronzy scales on fossa, usually with a more or less con-
spicuous linear pattern of golden scales surrounded by dark bronzy scales

. ae

2(1). Central part of anterior mesonotum with at least anterior inner dorsocentral
lines golden in addition to the acrostichal line. . .. co
- Central part of anterior mesonotum with only the acrostichal line golden .4

Berlin: Subgenus Howardina 15

3(2). Lateral mesonotal light line conspicuous anteriorly, broad, not interrupted at

or before scutal angle . . . . . . 2. Septemstriatus
Lateral mesonotal light line either completely absent Or very narrow or inter-
rupted at scutal angie; es ee ee eras
4(2). Posterior outer dorsocentral golden line distinct, linear . . 4. aureolineatus
Posterior outer dorsocentral golden line represented by a patch caudad of
scuta anes 68 bee ee ae ae ae oe ee ee
MALE GENITALIA

(5. Iquitos form unknown) ©

F; Subapical seta of claspette usually less than 0.7 of length of apical, latter not
reaching middle of sternomesal margin of sidepiece . Die:

. . 1. fulvithorax; 2. septemstriatus

Subapical seta of claspette usually subequal i in length to apical, latter extend-

ing beyond level of middle of sternomesal margin of sidepiece. . . . .2

2(1). Several subapical setae of claspette distinctly thickened . . 4. aureolineatus
None of the subapical setae of claspette thickened . . . . . 3. arborealis

PUPAE
(4. aureolineatus and 5. Iquitos form unknown)
iP Hair 1-II usually with 4 or more branches. . 1. fulvithorax; 2. septemstriatus
Haiw-i ll usually double @ 66456 sha a6 2 we ees ie Ay reals
LARVAE

(4. aureolineatus and 5. Iquitos form unknown)

L Hair 6-C usually double; 11-C double; most pecten teeth without basal den-
telee 2 ess . . 3. arborealis

Hair 6-C single; 11-C with more than 2 branches; all pecten teeth with basal
Genes: (ilies i ere ee ae ee TO a eee:

2(1). Hair O-P long, usually 14-branched (10-21); gills moderate, ha, unequal .
. 1. fulvithorax
Hair 0-P short, usually $-branched (4 0 lls long slendet and subequal.
SORE Oe BO8 .2. septemstriatus

1. Aedes (Howardina) fulvithorax (Lutz)
Figs.2,8,10,11

1904.Haemagogus fulvithorax Lutz, 1904:47,66,73. TYPE: Holotype ?, bridge Ipe-Arcado (Goias),
Brazil, date and collector not specified [LU].

1906.Taeniorhynchus palliatus Coquillett, 1901a:61. TYPE: Holotype , Trinidad, locality and
date not specified, F.W. Urich [USNM, 9140]. Synonymy with fulvithorax by Howard, Dyar
and Knab (1917:844).

1947.Aedes (Howardina) tachirensis Anduze, 1947:355-357. TYPE: Holotype ?, Las Mesas de

16 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Seboruco, Dto. Jauregui (Tachira), Venezuela, date and collector not specified [A] . Synony-
my with fulvithorax by Lane (1951:335).

Aedes (Howardina) fulvithorax of Bonne and Bonne-Wepster (1925:373-375); Dyar (1928:228-
229); Shannon (1931a:8,23,24,25; 1931b:142,143,146,161); Edwards (1932:155); Soper, Pen-
na et al (1933:576); Lane (1936b:186); Whitman and Antunes (1937:819-821); Lane (1939:
99); Floch and Abonnenc (1942a:5; 1942b:6); Anduze (1947:355); Lane (1951:335; 1953:
706-708); Horsfall (1955:413,471,483); Martinez and Prosen (1955:29-31); Vargas (1956:22);
Stone, Knight and Starcke (1959:174); Fauran (1961:29); Prosen, Carcavallo and Martinez
(1964:104); Forattini (1965:400); Cova Garcia, Sutil and Rausseo (1966a:57,285; 1966b:11,
73,141 ,169,333-334).

Aedes fulvithorax of Howard, Dyar and Knab (1912:65); Urich (1913:528); Howard, Dyar and
Knab (1917:844-846); Dyar (1918:73,80); Costa Lima (1930:255); Davis and Shannon (1931:
716,717,718); Martini (1931:206-208); Lane (1936a:131); Causey and Kumm (1948:475,476,
477); Causey and Santos (1949:478); Causey, Kumm and Laemmert (1950:302,307); Matheson
(1950:325); Delponte, Castro and Garcia (1951:239,240); Whitman (1951:263); Floch and
Fauran (1954:48); Aitken (1960:4); Belem Virus Lab (1964:101,103); Bates (1949:205,320).

Gualteria fulvithorax of Theobald (1907:552); Peryassu (1908:45,179-180); Theobald (1910: oe

Haemagogus fulvithorax of Lutz (1905:67); Dyar and Knab (1906a: 166,167). 3

Taeniorhynchus palliatus of Coquillett (1906b:24); Theobald (1910:432).

FEMALE. Wing: 2.97 mm. Proboscis: 2.34 mm. Forefemur: 2.16mm. Abdomen:
about 2.88 mm. Head: Scaling predominantly golden, a few dark scales present but
not forming distinct lateral patches; narrow median scales occupying an extensive tri-
angular area extending in the back to the outer erect scales and wide in front. Palpus
about 0.19 of proboscis. Torus light brown to yellow, usually darkened on mesal sur-
face. Thorax: Mesonotal integument yellowish to light brown; scaling usually entire-
ly golden to just before level of prescutellar space; numerous linear dark bronzy
scales along prescutellar rows, sometimes extending forward along dorsocentral bris-
tles, a few similar dark scales in front of prescutellar space, a variable patch of similar
dark scales above supraalar bristles, no dark scales on fossa; golden scales at scutal
angle not markedly broadened (fig.8). Median scutellar lobe with numerous broad
dark scales and a few golden in middle from base to apex; lateral lobes predominant-
ly with narrow golden scales. Pleural scaling silvery, with slight yellowish tinge; apn
predominantly with narrow scales, a few moderately broadened ones on lower and
posterior parts; ppn with sparse vestiture of short linear golden scales, usually with
only a few broadened scales in lower part. Abdomen: Median basal light patches usu-
ally present on tergites III, IV, light brown or golden; always present on V,VI, usually
with slight yellowish tinge.

MALE. Wing: 2.36 mm. Proboscis: 2.1 mm. Forefemur: 1.49 mm. Essentially sim-
ilar to female but lighter in color, dark scales of prescutellar area and midlobe of scu-
tellum less numerous; apn with large patch of broad scales on lower posterior sur-
face. Palpus distinctly shorter than proboscis.

MALE GENITALIA (fig.10). As figured; diagnostic characters as in the key. Ninth
tergite lobe not prominent, with about 5 bristles. Claspette with subapical seta not
thickened, less than 0.7 of apical, latter not reaching middle of sternomesal margin
of sidepiece. Clasper with 1,2 subapical hairs.

PUPA (fig.10). Abdomen: 3.24 mm. Trumpet: 0.38 mm; index 4.5. Paddle: 0.73
mm. Chaetotaxy as figured; hairs moderately to darkly pigmented. Apparently indis-
tinguishable from septemstriatus except perhaps by the longer trumpet. Readily sep-
arated from arborealis by hair 9-VI being minute, less than 0.5 of hair 2-VI, and more
numerous branches in hairs 1-II, 9-VII, VIII. Cephalothorax: Mesonotum moderately

Berlin: Subgenus Howardina 17

pigmented between trumpets; lateral areas and metanotum slightly darker. Hair 1(1,
1+2),° 201); 301); 4Q)s- 5Qa1-3),- 601), 7; 1-3)380- F3), 9Uy 10, 2-4) 11),
12(2, 1-4). Abdomen: Tergite I deeply pigmented except for usual membranous ar-
eas; tergite II uniformly moderately pigmented; tergites and sternites HI,[V with
moderate pigmentation but anterior border darker; rest of segments lighter. Segment
I: hair 1(10-14 primary branches, each with many secondary ones), 2(2, 1-4), 3(1),
4(4, 2-6), 5(3, 2-5), 6(1), 7(2, 1-3), 9(1), 10(1, present on majority of specimens but
sometimes absent on one side). Segment II: hair 0(1), 1(7, 4-13), 21, 1-2), 31, 1-3),
4(4, 3-5), 5(1, 1-2), 6(1, 1-2), 7(1, 1-4), 9(1). Segment III: hair 0(1), 1(2, 1-4), 2(1),
3(1)s.435 1-4; 501 2: 1-2),.601), 78:2-4).8@, 13), 90), 100, 1-2); LU, 140).
Segment IV: hair 0(1), 1(1, 1-2), 2(1), 3(3, 1-4), 4(2, 1-3), 5(1), 6(1), 7(3, 2-3), 8@,
2-3), 9(1), 10(1, 1-3), 11(1), 14(1). Segment V: hair 0(1), 11, 1-2), 2(1), 31, 1-2),
4(2, 1-4), 5(1), 6(1), 7(3, 2-4), 8(2, 1-3), 9(1), 10C1, 1-2), 111), 141). Segment VI:
hair O(1, 1-2), 1(1), 2(1), 3(1), 4, 1-2), 51), 6(1), 71, 1-2), 8@2, 1-2), 9(1), 101),
11(1), 14(1). Segment VII: hair 0(1), 1(1), 21), 3), 41), 5(1), 6(@, 1-4), 70), 8@,
1-3), 9(4, 3-5), 10(1), 111), 14(1). Segment VIII: hair 0(1), 4(1), 9(7, 5-8), 14(1).
Paddle: Uniformly lightly pigmented, midrib darker; external margin slightly ser-
rated; hair 1-P slender; male genital lobe extending to about 0.37 of paddle, female
genital lobe to 0.29.

LARVA (fig.11). Head: 0.89 mm. Siphon: 0.7 mm. Anal Saddle: 0.27 mm. Both
hairy and nonhairy forms known. Very similar to septemstriatus from which it can
be separated apparently by the longer, more profusely branched hair 0-P and the
broad, short unequal gills. Head: Pigmentation moderate except around mouthparts
where it is deeper; integument smooth. Mental plate with a strong median tooth and
11(11,12) distinct teeth on either side, often a smaller detached tooth at base. Hair
O(1), 1(1), 3(1), 4(13, 10-16), 5(1), 6(1; moderate), 7(3, 2-4), 8(1), 9(2, 1-2), 10(1),
11(12, 11-16; stellate), 12(3, 2-4), 13(1), 14(4, 3-8), 15@, 2-4). Antenna: Length
about 0.48 of head; pigmentation deeper than on head capsule; all hairs single; hair |
at 0.5 from base. Thorax: Apices of long hairs attenuate; barbs present on long and
stellate hairs. Prothorax: hair 0(14, 10-21), 1(4, 3-4), 2(1), 3(7, 4-8), 444, 4-9), 5@,
2-4), 6(1), 7(3, 2-3), 8(11, 8-18), 9(2, 1-5), 10(1), 11@, 1-4), 121), 14(21, 17-28).
Mesothorax: hair 1(5, 3-8), 2(1), 3(1), 4(1), 5(1), 6(5, 4-6), 7(1), 865, 4-7), 965, 4-7),
10(1), 11(2, 1-3), 12(1), 13(15, 12-23), 14(9, 8-16). Metathorax: hair 1(5, 4-12),
2(1), 3(4, 3-6), 4(6, 5-13), 5(5, 4-13), 6(1), 7(5, 4-8), 809, 8-22), 9(4, 3-5), 10(1),
11(1, 1-2), 12(1), 13(11, 8-18). Abdomen: Stellate hairs with branches of unequal
length. Segment I: hair 1(8, 7-13), 2(10, 4-10), 3(1), 4(6, 5-6), 5(12, 10-13), 6(4,
3-4), 7€1, 1-2), 9(6, 5-8), 10(1), 11(12, 6-13), 13(8, 5-12). Segment II: hair 0(1),
1(1:0;:3-441); 209; 6-10), 31), 465, 3-8), 5(10,7-11); 66, 3-4), 73, 2-7), (1), 9(8,
5-10), 10(1), 111), 12(1), 13(8, 5-14). Segment III: hair 0(1), 1(4, 2-8), 2(7, 7-9),
3(1), 4(1), 5(9, 6-12), 6(2), 7(6, 5-10), 8(1), 9(9, 6-13), 10(1), 11(1), 12(1), 13,
4-12), 14(1). Segment IV: hair 0(1), 1(5, 2-9), 2(8, 7-9), 3(1), 4(1), 5(9, 6-11), 6(2),
7(5, 5-8), 8(1), 9(8, 7-11), 10(1), 11(1), 121), 134, 2-10), 14(1). Segment V: hair
0(1), 1(3, 2-9), 2(7, 6-9), 3(2, 1-2), 4(1), 5(6, 6-9), 6(2), 7(5, 5-8), 8(1), 9(8, 6-10),
10(1), 1101), 12(1), 13(5, 2-11), 14(1). Segment VI: hair 0(1), 1(5, 2-10), 2(7, 6-9),
3(1), 4(1), 5(7, 4-11), 6(1, 1-2), 7(4, 3-7), 8(1, 1-2), 9(4, 4-11), 10(1), 11€1), 121),
13(6, 5-13), 14(1). Segment VII: hair 0(1), 1(3, 2-8), 2(7, 6-10), 3(1), 4(1), 5(4, 3-8),
6(3, 2-3), 7(1), 8(5, 4-7), 9(4, 3-6), 10(1), 11(1), 121), 134, 3-10), 14(1). Segment
VII. Comb scales 20(19-24), in a patch of 3 irregular rows; individual scale elongate,
rounded apically and evenly fringed with short spinules; hair 0(1), 1(5, 4-10), 2(1),
3(7, 5-8), 4(1), 5(6, 5-11), 14(1). Siphon: Index about 2.25-2.3; darkly pigmented

18 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

and with imbricate sculpturing throughout. Pecten extending to about 0.34-0.38;
with 13 teeth (10-15), usually all with 1-3 basal denticles, terminal tooth not mark-
edly enlarged. Hair 1(3, 3-4), 2(1), 3-5 not visible, 6(1), 7(1), 8(@2, 1-3), 9(1). Anal
Segment: Saddle darkly pigmented, with uniform imbricate sculpturing. Gills broad
and short, unequal, ventral about 0.5 of dorsal. Hair 1(3, 2-7), 2(4, 3-5), 3(1), 4a(4,
3-5), 4b(3, 3-4), 4c(3), 4d(3, 3-4), 4e(3, 3-4), 4f(3, 2-4).

SYSTEMATICS. Adults of fulvithorax show considerable variation in thoracic or-
namentation in different areas of the reported wide distribution of this species but
are readily separated from the other species of the group by the much greater amount
of golden scaling on the mesonotum in both males and females. The above descrip-
tions of all stages are based entirely on material from Trinidad, the type locality of
palliatus. The single specimen I have seen from Venezuela, a female paratype of tach-
irensis, is indistinguishable from the Trinidad material. In these 2 northern popula-
tions, the integument of the thorax is light yellowish tan and the mesonotal scaling
is bright yellow golden without any indication of dark scales on the fossa or anterior
dorsocentral lines. The Surinam specimen agrees with this description also. The
French Guiana material is similar in scaling but the thoracic integument is brown. As
one proceeds south along the Atlantic lowlands of Brazil (Para, Bahia and Rio de Ja-
neiro) the mesonotal golden scaling becomes darker, more coppery and the integu-
ment tends to become darker brown; however, in all these populations the fossa is
without dark bronzy scales. At higher elevations in Brazil (Sao Paulo, Goias, Matto
Grosso), the darkening of the integument and scaling becomes even stronger and
there is a tendency for the appearance of a few dark bronzy scales in the fossa and
along the posterior dorsocentral line. The single female from Iquitos resembles the
interior populations of Brazil but has more extensive dark-scaled patches on the head
than any other one I have seen.

In the scanty material of immature stages available from populations other than
palliatus, no striking differences are apparent.

No clear-cut separation can be made of fulvithorax from septemstriatus at this
time in the pupal stage but the larvae of the 2 species are apparently distinct on the
_ characters given in the key and diagnoses.

BIONOMICS. The immature stages of fulvithorax have been found only in tree-
holes. The adults are primarily diurnal in activity but a few specimens have been col-
lected at night.

Females feed readily on experimental animals in the laboratory. They have been
investigated as possible vectors of yellow fever virus (Davis and Shannon 1931; Ait-
ken 1960). The virus will survive and apparently develop in these mosquitoes as
shown by successful infection of rhesus monkeys by inoculation of experimentally
infected fulvithorax but all attempts to transmit the virus by bite have been negative.

DISTRIBUTION (fig.2). Lowlands to moderate elevations in Venezuela (Maracai-
bo and Orinoco drainages), Trinidad, Guyana, Surinam, French Guiana, Brazil (Ama-
zon, Paraguay and Parana drainages, and states of Espirito Santo and Rio de Janeiro)
and Peru (Amazon drainage). Material examined: 152 specimens; 20 6, 62 9, 29 pu-
pae, 41 larvae; 30 individual rearings (4 pupal, 20 larval, 6 incomplete).

BRAZIL. Bahia: Piraja, 2 Apr 1929, [C. Davis and] R.C. Shannon, 2 Ip, 2 6, 1 6 genitalia, 5 L,
6 1; 1938, 1 6d; 1 1 [USNM]. Espirito Santo: Santa Teresa, Apr 1940, 2 ° (Martinez and Prosen,
1955:31; MEPRA). Valle do Chanaan, 173 @ (Davis and Shannon 1931:715; Whitman and An-
tunes 1937:819-820). Goias: Anapolis, Sept 1936, 1 6, 1 9 [USNM]. Matto Grosso: Guiaba, Feb
1935, 2 2; Mar 1935, 3 ? [USNM]. Maracaju, July 1937, 1 9; Nov 1937, 1 9; Dec 1937, 8 9;
3 2 [USNM]. Pocinho (Lane 1936b:186). Minas Gerais: Passos, 1945-47, 23A (Causey and Kumm

Berlin: Subgenus Howardina 19

1948:475,477); 1840A (Causey, Kumm and Laemmert 1950:302,307); 386A (Causey and Santos
1949:478). Para: Belem, IAN forest, 105 9; IAN and Utinga forests, 9 2; Utinga forest, 9 ? (Belem
Lab Report 1964:101,103,109,114). Belterra, June 1938, 1 2 [USNM]. Ilha de Marajo, 126 ?
(Kumm and Novis 1938:502). Urucurituba, 18 Mar, C.H. Townsend, 1 9 [USNM] . Rio de Janeiro:
Apr 1938, 4 9 [USNM]. Sao Paulo: Guariroba; Nhandeara; Nipoan; Planalto; Villa Poloni (Lane
1936a:131). State not specified: 1 3,1 [USNM].

FRENCH GUIANA. Baduel (Fauran 1961:29). Balourou, Crique Anguille, Galion (Floch and
Abonnenc 1947:10). Placer Isnard, June 1940, E. Abonnenc, 1 @ [UCLA] (Floch and Abonnenc
1947:10; Fauran 1961:29). Rorota, 8 May 1941, E. Abonnenc, 1 ¢ [UCLA] (Floch and Abon-
nenc 1947:10). Souvenir (Floch and Abonnenc 1947:10). Locality not specified, 1944, 1 d, 1
[USNM]. |

GUYANA. Mazaruni, 25 June 1936, 1 2 [USNM]; 6 July 1936, Leslie, 1 d (KO 208A-1)
[UCLA].

PERU. Iquitos, Mar-Apr 1931, R.C. Shannon, 1 9 [USNM].

SURINAM. Moengo, 19 Mar 1946, H.H. Stage, 1 L [USNM]. Locality not specified, 1 (36-
69) [ITH].

TRINIDAD. Nariva: Charuma forest, 27 Aug 1964, treehole, A. Guerra, 1 lp? (TR 634-108)
[UCLA]. St. Andrew: Mt. Harris, elev. 250 ft, 16 July 1964, treehole, F. Powdhar (TR 566), 1 Ipd
(566-102), 1 Ip? (566-108) [UCLA]. St. George: Arima, Aripo Valley, elev. 500 ft, 16 Apr 1964,
bamboo, A. Guerra (TR 319), 1 lpd (319-148), 2 Ip? (319-139,149), 1 p? (319-111), 2 IP (319-
128,137), 1 L; Cumaca road, elev. 500 ft, 14 Jan 1965, A. Guerra, 1 2 (TR 943); Verdant Vale,
elev. 1000 ft, 10 Sept 1964, treehole, A. Guerra, 2 lp? (TR 680-102,103) [UCLA]. Monos Island,
Grand Fond, 17 May 1964, treehole, A. Guerra (TR 410), 2 Ipd (410-106,107), 1 Ip? (410-111);
1 Aug 1964, bamboo, R. Martinez (TR 584), 1 lp? (584-11), 1 pd (584-102); 2 Aug 1964, bam-
boo, R. Martinez (TR 591), 1 lpd (591-109), 4 lp? (591-113,114,119,121), 1 pd (591-108), 1 pe
(591-107), 1 L [UCLA]; 17 Nov 1905, F.W. Urich, 3 6, 1 ?; June 1906, 2 Ip? (20-2,3), 1 12 (20-1),
1 lp (20-4), 4 9? [USNM]. San Rafael, Arena forest, Oct 1965, 2 9 (TR 1432), Dec 1965,3 9 (TR
1441) [UCLA]. Port of Spain, 9 Sept, 1 9; 1 9 [USNM] . County not specified: 26 Aug 1945, 1 Ip;
9 Sept 1945, 1 ? [USNM].

VENEZUELA. Bolivar: Distrito Heres (Anduze 1947:355). Rio Orocopiche (Cova Garcia, Su-
til and Rausseo 1966b:73). Tachira: Chururu, Distrito Uribante (Anduze 1947:357). Las Mesas
de Seboruco, Distrito Jauregui, June 1944, P.J. Anduze, 1 9 [USNM] (Anduze 1947:357; Cova
Garcia, Sutil and Rausseo 1966b:73).

2. Aedes (Howardina) septemstriatus Dyar & Knab
Figs.2,8,12,13

1907.Aedes septemstriatus Dyar and Knab, 1907:10. TYPE: Holotype ? (8), Bluefields, Nicaragua,
W.F. Thornton [USNM, 10144].

Aedes (Howardina) septemstriatus of Bonne and Bonne-Wepster (1925 :366-367); Dyar (1925:141,
148; 1928:232-233); Shannon (1931:142,147); Edwards (1932:155); Lane (1939: 100); Floch
and Abonnenc (1942a:6; 1942b:5; 1944:4; 1947:10); Martinez and Prosen (1955:27-29); Reyes
(1955:74,89); Trapido, Galindo and Carpenter (1955:530,537,538); Trapido and Galindo
(1957:122,125); Stone, Knight and Starcke (1959:174); Fauran (1961:29); Belkin, Schick and
Heinemann (1965:42); Prosen (1965:42); Forattini (1965:239,242,403-404).

Aedes septemstriatus of Theobald (1910:485); Howard, Dyar and Knab (1917:846-847); Dyar
(1918:73); Komp (1936:63); Antunes (1937a:16; 1937b:76); Arnett (1949:238); Galindo, Tra-
pido and Carpenter (1950:102,104,105 108,112,113); Carpenter and Peyton (1952:681); Galin-
do, Carpenter and Trapido (1955:163); Reyes (1955:75); Komp (1956:39); Stone and Knight
(1956:225); Blanton and Peyton (1958:31-32).

Aedes arborealis of Antunes (1937b:76); Reyes (1955:74,89); Boshell-Manrique and Osorno-Mesa
(1944:177).

20 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

FEMALE. Wing: 3.15 mm. Proboscis: 2.53 mm. Forefemur: 2.01 mm. Abdomen:
about 2.8 mm. Head: A large patch of broad dark scales on each side of a narrow me-
dian stripe of narrow golden scales, this stripe narrowed caudad. Palpus about 0.14
of proboscis. Torus dark brown. Thorax (fig.8): Mesonotal integument dark brown;
mesonotum with a striking pattern of golden lines on a field of dark bronzy scales,
latter sometimes with a few scattered golden scales; golden scaled lines as follows,
narrow acrostichal bifurcating at prescutellar space, narrow inner dorsocentral not
always reaching prescutellar space, outer dorsocentral arising at about level of scutal
angle where it is broad and extending as a narrow line to scutellum; lateral marginal
line broad, golden, interrupted in front of wing root, golden scales at scutal angle
broader than in fulvithorax; fossa predominantly with dark bronzy scales, but some-
times with some scattered golden scales; space dorsad of supraalars with a large con-
spicuous patch of dark bronzy scales. Scutellar ornamentation as in fulvithorax. Pleu-
ral scaling denser and more silvery than in fulvithorax; scales of apn predominantly
broad and very dense; ppn scale patch more extensive than in fulvithorax and with
distinct broad scales in lower part. Abdomen: Median basal tergal light patches essen-
tially as in fulvithorax except that scales on V,VI more silvery, without distinct yel-
low tinge.

MALE. Wing: 2.7 mm. Proboscis: 2.57 mm. Forefemur: 1.84 mm. Mesonotal or-
namentation markedly different from female; anterior part predominantly with gold-
en scales as in fulvithorax except for very narrow inconspicuous lines of dark bronzy
scales on anterior dorsocentral lines and on each side of moderately broad acrostichal
line, fossa usually without any bronzy scales; posterior part of mesonotum with
more extensive pale scaling than in female but upper supraalar area with a distinct
large patch of bronzy scales. Head with smaller dark patches than in female; palpus
longer than in fulvithorax, subequal in length to proboscis.

MALE GENITALIA (fig.12). As figured; diagnostic characters as in the key. Ex-
tremely similar to fulvithorax but apparently with a more prominent IX tergite lobe
with 8,9 bristles, and clasper with 2,3 longer subapical hairs.

PUPA (fig.12). Abdomen: 2.97 mm. Trumpet: 0.48 mm; index 4.3-4.5. Paddle:
0.76 mm. Only nonhairy form known. Extremely similar to and apparently indistin-
guishable from fulvithorax ; separated from arborealis by the same characters as fulvi-
thorax. Chaetotaxy as given below; hairs moderately to darkly pigmented. Cephalo-
thorax: Mesonotum lightly pigmented; metanotum moderately pigmented; trumpet
deeply pigmented and sharply contrasting. Hair 1(1), 2(1, 1-2), 3(1), 4C1, 1-2), 5(2,
1-3), 6(1), 7@2, 1-2), 8(1, 1-2), 9(1), 10(2, 1-3), 11(1), 12(2, 1-3). Abdomen: All ab-
dominal segments lightly pigmented, first 3 segments slightly darker; anteromedian
area of sternites III and IV with darker patches; integument imbricate. Segment I:
hair 1(9-15 primary branches subequal in length and each with many secondary
branches), 2(1), 3(1), 4(5, 2-6), 5(3, 2-5), 6(1), 7(1), 9(1), 10(1), and 11(1, found
occasionally). Segment II: hair 0(1), 1(5, 2-7), 2(1), 3(1), 4(4, 2-6), 5(1, 1-2), 6(1),
7(1, 1-2), 8(1, found occasionally), 9(1). Segment III: hair 0(1), 1(1, 1-2), 2(1), 3(1),
4(2, 1-3), 5(1), 6(1), 7(3, 1-4), 8(2, 1-3), 9(1), 10(1, 1-2), 11(1), 14(1). Segment IV:
hair O(1), 1(1), 2(1), 3@, 1-3), 4(1, 1-2), 5(1), 6(1), 7(@2, 1-3), 8(@2, 1-3), 9(1), 10C1,
1-2), 11(1), 14(1). Segment V: hair 0(1), 1(1), 2(1), 3(1, 1-2), 4(2, 1-3), 5(1), 6(1),
7(3, 1-4), 8(2, 1-2), 9(1), 10(1), 11€1), 14(1). Segment VI: hair 0(1), 1(1), 2(1), 3(1),
4(1), 5(1), 6(1), 701), 8(2, 1-2), 9(1), 101), 11(1), 14(1). Segment VII: hair 0(1),
I(1), 21), 3(1), 41), 51), 6(1, 1-2), 7(1), 8(2, 1-2), 9(3, 2-4), 10(1), 11(1), 14(1).
Segment VIII: hair 0(1), 4(1), 9(5, 4-6), 14(1). Paddle: Lightly pigmented, midrib
slightly darker; apex not emarginated and with minute denticles; hair 1(1, 1-2). Male

Berlin: Subgenus Howardina ) 21

genital lobe extending to 0.35 of paddle, female genital lobe to 0.31.

LARVA (fig.13). Head: 0.9 mm. Siphon: 0.78 mm. Anal Saddle: 0.32 mm. Only
nonhairy form known. Very similar to fulvithorax but apparently distinct from the
latter in the short hair O-P with few branches and gills long, slender and subequal.
Chaetotaxy as given below; hairs moderately pigmented. Head: Pigmentation moder-
ate except for areas around mouthparts where it is deeper. Mental plate with a strong
median tooth and 12 distinct teeth (10-12) on each side, often a smaller detached
tooth at base. Hair 0(1). 1(1), 3(1), 4(15, 9-16), 5(1), 6(1), 7(4, 2-6), 8(1), 9(1),
10(1), 116, 3-7), 12(2, 1-3), 13(1), 14@2, 2-3), 15(2, 1-2). Antenna: Length about
0.41 of head; shaft uniform up to middle and then gradually narrowed distally; uni-
formly moderately pigmented, without spicules; all hairs single; hair 1 at about 0.5
from base. Thorax: Apices of long hairs attenuate; integument without spicules; stel-
late hairs with few branches. Prothorax: hair 0(5, 4-6), 1(3), 2(1), 3(3, 3-5), 4(3, 2-
4), 5(3, 3-4), 6(1), 7(3, 2-3), 8(5, 5-8), 9(1, 1-2), 10(1), 11(2, 1-2), 12(1), 14(14, 10-
18). Mesothorax: hair 1(3, 3-4), 2(1), 3(1), 4(1), 5(1), 6(5, 4-5), 7(1), 8(5, 5-6), 9(5,
5-7), 10(1), 111, 1-2), 12(1), 13(7, 3-8), 14(5, 3-5). Metathorax: hair 1(4, 3-5), 2(1),
3(3, 3-5), 4(3, 3-5), 5(5, 3-6), 6(1), 7(5, 5-7), 8(6, 4-7), 9(4, 3-5), 10(1), 11(1, 1-2),
12(1), 13(4, 3-6). Abdomen: Stellate hairs with fewer branches than in fulvithorax.
Segment I: hair 1(5, 4-5), 2(6, 3-6), 3(1), 4(4, 2-8), 5(6, 5-7), 6(4, 3-4), 7(1), 9(3, 3-
5), 10(1), 11(6, 5-8), 13(2, 2-5). Segment II: hair 0(1), 1(3, 2-4), 2(6, 5-8), 3(1), 4(3,
2-5), 5(5, 4-6), 6(3), 7(2, 1-3), 8(1), 9(4, 3-5), 10(1), 11(1), 12(1), 133, 2-5). Seg-
ment III: hair 0(1), 1(2, 1-4), 2(7, 4-7), 3(1), 4(1), 5(5, 3-5), 6(2), 7(4, 2-5), 8(1),
9(6, 4-7), 101), 11(1), 12(1), 13(@2, 2-3), 14(1). Segment IV: hair 0(1), 1(2), 2(6, 5-
8), 3(1), 4€1), 5(4, 4-6), 6(2), 7(3, 3-5), 8(1), 9(7, 4-7), 10(1), 11(1), 1201), 13Q, 2-
3), 14(1). Segment V: hair 0(1), 1(2, 1-2), 2(7, 5-8), 3(1), 4(2, 1-2), 5(5, 3-5), 6(2),
7(3, 3-4), 8(1), 9(6, 4-7), 10(1), 11(1), 12(1), 13@, 1-3), 14(1). Segment VI: hair
O(1), 1(2), 2(6, 6-7), 3(1), 4(@2, 1-2), 5(4, 3-5), 6(1, 1-2), 773, 3-5), 8(1), 9(4, 3-5),
10(1), 11(1), 12(1), 13(2, 2-4), 14(1). Segment VII: hair 0(1), 1(2), 2(6, 6-7), 3(1),
4(1), 5(3, 2-4), 6(2, 1-4), 7(1), 8(3, 2-5), 9(3, 3-4), 10(1), 111), 12(1), 13(2, 2-5),
14(1). Segment VIIT: Comb scales usually 20(13-27), in 3 irregular rows, elongate,
rounded apically and evenly fringed; hair 0(1), 1(3, 2-3), 2(1), 3(5, 4-6), 4(1), 5(4, 3-
4), 14(1). Siphon: Index about 2.2-2.3; darkly pigmented and imbricate. Pecten ex-
tending to about 0.4; usually 15 teeth (12-15), smaller at base and gradually becom-
ing larger distally, all with 1-3 basal denticles. Hair 1(3, 2-3), 2(1), 3-5 not visible,
6(1), 7(1), 8(@2, 1-3), 9(1). Anal Segment: Saddle darkly pigmented and uniformly
imbricate. Gills moderately long, narrow and subequal. Hair 1(2), 2(4, 3-5), 3(1),
4a(3, 3-4), 4b(2, 2-3), 4c(2, 2-3), 4d(2, 2-3), 4e(2, 2-3), 4£(2).

SYSTEMATICS. In the present interpretation, septemstriatus appears to be re-
stricted to areas of rain forest, chiefly in the lowlands, in Nicaragua, Panama and Co-
lombia. In the past it has been confused with arborealis and aureolineatus but it ap-
pears to be more closely related to fulvithorax on the basis of male genitalia and im-
mature stages.

This species has a more marked sexual dimorphism in adult ornamentation than
any other member of the group, the male approaching the condition in both sexes of
fulvithorax and the female having a striking arrangement of golden and dark bronzy
scales in lines on the mesonotum. In the area of overlap of the 2 forms in Colombia,
fulvithorax and septemstriatus are very strongly differentiated and there is no indi-
cation of intergradation; for this reason septemstriatus is considered here to be a dis-
tinct species.

There is considerable variation in the ornamentation of the adults of septemstria-

Ze Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

tus but the available material is too scanty to determine whether this is entirely in-
dividual, ecological or geographical. However, it appears that in Colombia the light
scaling is more restricted than in Panama and Nicaragua and particularly that the sex-
ual dimorphism is not as marked. The above descriptions are based on individually
associated material from Panama.

Although the adults of septemstriatus are usually readily separated from all other
members of the group on the diagnostic characters in the key and descriptions, I
have not been able to find any reliable characters to separate it in the pupal stage
from fulvithorax. The larvae of the 2 species are also very similar and it is possible
that some of the characters used to differentiate them are not reliable because only
the nonhairy form of septemstriatus has been seen.

BIONOMICS. The immature stages of septemstriatus have been found in treeholes
and broken bamboo. In one instance they were associated with 3 species of Haema-
gogus, Aedes (F.) terrens, 2 species of Culex and a species of Sabethes (Galindo, Car-
penter and Trapido 1955). The adults are diurnal, primarily sylvan, and are found
mainly at the forest floor and not in the canopy (Trapido, Galindo and Carpenter
1955; Trapido and Galindo 1957). Other published reports on the bionomics of sep-
temstriatus (Kumm and Novis 1938; Causey and Santos 1949; Floch and Fauran
1954; Belem Virus Lab 1964) do not apply to this species and should probably be
attributed to arborealis.

DISTRIBUTION (fig.2). Lowland to moderate elevations in Colombia (Upper Ori-
noco drainage), Costa Rica (Pacific side), Nicaragua (Atlantic side) and Panama (At-
lantic and Pacific sides). Material examined: 151 specimens; 44 6, 57 °, 42 pupae, 8
larvae; 12 individual rearings (9 pupal, 3 larval).

COLOMBIA. Meta: Restrepo, 12 July 1935, W.H.W. Komp, 2 6 [USNM], 2 ? (KO 209A-9), 1°
(KO 209A-11) [UCLA]; 22 July 1935, W.H.W. Komp, 9 6, 3 , 24; July 1935, 1 6 genitalia, 1 L;
22 Aug 1935, W.H.W. Komp, 1 6 [USNM], 1 ? (KO 206A-1), 4 6, 8 2 (KO 206A-7), 1 6 (KO
206A-13) [UCLA]; 2 May 1942, W.H.W. Komp, 2 2; 11 [USNM]. Villavicencio, 16 July 1943,
W.H.W. Komp, | 9; 25 Oct 1943, M. Bates, 3 9; 1944, M. Bates, 1 p, 1 1 [USNM].

COSTA RICA. Puntarenas: Esquinas, elev. 300 ft, 6 June-30 Nov 1950, 1 A (Trapido, Galindo
and Carpenter 1955:530).

NICARAGUA. Bluefields, W.F. Thornton, 2 9? topotypes [USNM].

PANAMA. Bocas del Toro: Almirante, Bocas-Chiriqui Road, 2 May 1963, A. Quinonez (PA
301), 3 Ipd (301-106,109,112), 5 pd (301-102,105 108,110,111), 4 p? (301-101,103,104,109),
29 P. Chiriquisito, Trail Caldera, 5 Nov 1965, P. Orjuela (PA 1004), 5 9 [UCLA]. Canal Zone: Em-
pire, 11 Feb 1926, J.B. Shropshire, 1 ¢. Fort Randolph, 27 July 1926, D.P. Curry, 1 6. France
Field, 20 Feb 1926, J.B. Shropshire, 6 6, 19 9. Mojinga Swamp, 1 A (Blanton and Peyton 1958:
32). Paja (=Nueva Emperador), 13 May 1941, G.B. Fairchild, 5 d, 1 2. Summit, 30 May 1938,
W.H.W. Komp, 1 ?; 7 May 1943, W.H.W. Komp, 1 6; 15 May 1945, W.H.W. Komp, 1 6; 1 ?
[USNM] . Chiriqui: Chorcha, elev. 300 ft, 12 June-11 Dec 1950, 1 A (Trapido, Galindo and Car-
penter 1955:530). David, C.S. Ludlow, 1 @ (Dyar 1923:148) [Army Medical Museum]. Cocle:
Tucue, elev. 1000 ft, 12 June-14 Dec 1950, 14 A (Trapido, Galindo and Carpenter 1955:530).
Colon: Candelaria, elev. 200 ft, F4-27 Aug 1950, 5 A (Trapido, Galindo and Carpenter 1955:530).
Panama: Arraijan, elev. 600-800 ft, 5 June-29 Dec 1950, 2 A. Campana, elev. 2000 ft, 14 June-6
Dec 1950, 12 A. Campanita Arriba, elev. 600 ft, 24 Aug-4 Dec 1950, 14 A. La Victoria, elev. ©
400-1200 ft, 7 Feb-28 Dec 1950, 1 A (Trapido, Galindo and Carpenter 1955:530). Pacora area,
10 2 (Galindo, Trapido and Carpenter 1950:548,551). San Jose Island: 26 July 1944, W.H.W.
Komp, 1 9 [USNM].

3. Aedes (Howardina) arborealis Bonne-Wepster & Bonne
Figs.2,8,14,15

Berlin: Subgenus Howardina Pe

1920.Aedes arborealis Bonne-Wepster and Bonne, 1920:178. TYPE: Lectotype 6 (3893) with as-
sociated larval and pupal skins (M 53) and genitalia slide (M 41), both slides marked BB 655,
Dam, Surinam, Jan 1919, J. Bonne-Wepster and C. Bonne [ITH; selection of Belkin 1968].

Aedes (Howardina) arborealis of Bonne and Bonne-Wepster (1925:377-380); Dyar (1928:23 1-232);
Edwards (1932:155); Lane (1936b:186; 1939:99); Floch and Abonnenc (1944:41; 1947:10);
Lane (1953:708-710); Horsfall (1955:471); Stone, Knight and Starcke (1959:173); Floch and
Fauran (1954:48); Fauran (1961:29); Belkin, Schick and Heinemann (1965:63); Forattini
(1965:400); Stone (1967:208).

Aedes septemstriatus of Kumm and Novis (1938:502,507,510-511); Causey and Santos (1949:
478); Floch and Fauran (1947:10; 1954: 48); Fauran (1961: 29); Belem Virus Lab (1964:101,
103,109,112).

FEMALE. Wing: 2.8 mm. Proboscis: 2.45 mm. Forefemur: 1.93 mm. Abdomen:
about 2.63 mm. In most respects intermediate between fulvithorax and septemstria-
tus but differing from both in the following. Head: Median line of golden scales very
narrow, these scales linear; submedian dark patches distinct but smaller than in sep-
temstriatus. Thorax (fig.8): Mesonotum predominantly with golden yellow scales an-
teriorly; ornamentation superficially very similar to 9 of septemstriatus except that
golden acrostichal line narrower, fossa predominantly with bronzy scales, and mar-
ginal golden area interrupted in front as well as caudad of scutal angle; caudal part
of mesonotum with dark scaling more extensive than golden; upper supraalar patch
of dark scales conspicuous. Scutellum with a conspicuous median line of golden
scales. Scales of apn, ppn, and ppl golden, remaining pleural scales silvery; apn with
numerous dense broad scales; ppn patch restricted, without broad scales. Abdomen:
Median basal tergal patches as in septemstriatus on VI,VII but apparently small and
yellowish on IV,V.

MALE. Wing: 2.8 mm. Proboscis: 2.62 mm. Forefemur: 1.75 mm. Mesonotal or-
namentation essentially as in the female but golden scales more numerous; acrostichal
golden line broad; paraacrostichal dark-scaled line very narrow; outer dorsocentral
dark line evident only posteriorly; fossa with small patch of dark bronzy scales; lower
part of ppn with a few broad scales. Head: Without distinct dark patches, with only
a few dark scales. Palpus subequal to proboscis in length, with numerous pale scales
on lower basal surfaces of segments 3 to 5.

MALE GENITALIA (fig.14). As figured; diagnostic characters as in the key. In
general similar to fulvithorax except for the following. Ninth tergite lobe indistinct,
with 6,7 strong bristles. Claspette with subapical seta subequal to apical but not
markedly thickened; other setae much longer than in fulvithorax, apical seta extend-
ing beyond middle of sternomesal margin of sidepiece. Clasper with about 5 setae in
distal half.

PUPA (fig.14). Abdomen: 3.24 mm. Trumpet: 0.44 mm; index 3.3-3.4. Paddle:
0.65 mm. Only nonhairy form known. Readily differentiated from fulvithorax and
septemstriatus by fewer branches in hairs 1-II, 9-VII, VIII and broader paddle. Chae-
totaxy as given below; hairs moderately to darkly pigmented. Cephalothorax: Meso-
notum between trumpets moderately pigmented, darker laterad; metanotum moder-
ately pigmented; trumpets darkly pigmented and strongly contrasting. Hair 1(1), 2(1),
3(1), 4(1, 1-2), 5C1, 1-2), 6(1, 1-2), 7(1, 1-2), 8(2, 2-3), 9(1), 10(2), 11(1), 12(1). Ab-
domen: Tergite I moderately pigmented except for usual light membranous areas;
other segments moderately pigmented and integument imbricate; pigmentation pro-
gressively lighter caudad. Segment I: hair 1(5-7 main branches, subequal in length
and each with many secondary branches), 2(1), 3(1), 4(2, 2-4), 5(1), 6(1), 7(1, 1-2),

24 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

9(1). Segment II: hair 0(1), 1(2, 1-4), 21, 1-2), 3(1), 4¢4, 1-4), 5(1), 6(1), 71, 1-2),
9(1). Segment III: hair 0(1), 1(1), 21), 31, 1-2), 4C1, 1-3), 51), 6(1), 7@, 1-4), 8@,
1-2), 9(1), 101, 1-2), 111, 1-2), 14(1). Segment IV: hair 0(1), 1(1), 2(1), 3@, 2-3),
41, 1-3), 5(1), 6(1), 71, 1-2), 8C1, 1-2), 9(1), 10€1), 11€1), 141). Segment V: hair
O(1), 101), 201), 301), 4(@, 1-3), 3(1),, 61), 7G, 1-2), 8(@2, 122), 901), 10d), 110),
14(1). Segment VI: hair 0(1), 1(€1), 2(1), 31), 4€1), 51), 6(1), 7(1), 8C1, 1-2), 9(1),
101), 11(1), 14(1). Segment VII: hair 0(1), 1(1), 2(1), 3(1), 4€1), 5C1), 6(1), 7Q1),
8(1), 9(2, 1-2), 10(1), 111), 141). Segment VUIT: hair 0(1), 4(1), 9(3), 14(1). Pad-
dle: Ovoid; lightly pigmented, midrib dark; apex slightly produced; hair 1 single;
male genital lobe extending to 0.6 of paddle, female genital lobe to 0.34.

LARVA (fig.15). Only nonhairy form known. Readily differentiated from fulvi-
thorax and septemstriatus by the following. Head hairs 6,11-C usually double; stel-
late hairs with heavier spikes; comb scales usually more numerous (27-39), in 4,5
rows, siphon index about 2.0; 10-11 pecten teeth without basal denticles, the distal
teeth greatly elongate; anal gills shorter than saddle, broad and subequal.

SYSTEMATICS. As indicated in the diagnosis above, arborealis is somewhat inter-
mediate between fulvithorax and septemstriatus in the ornamentation of the adults.
Its larva and pupa are quite distinct from these species but may be similar to the un-
known immature stages of aureolineatus which appears to be closely related to ar-
borealis.

In the present interpretation, arborealis has a more restricted distribution than for-
merly reported, being confined to the Guianas and the lower Amazon. Former re-
ports of arborealis from Colombia pertain to septemstriatus and all the specimens
from the lower Amazon identified as septemstriatus in USNM appear to be arborealis.

The above descriptions are based on the type series from Surinam. Specimens
from Guyana and French Guiana agree in all respects with the Surinam material. The
adults from the lower Amazon are usually considerably darker, with more extensive
dark bronzy scaling on the mesonotum, but there is a great deal of variation in this
character and some specimens are indistinguishable from the material from French
Guiana. I consider all these populations as members of a single species.

A single specimen of aureolineatus from Para without specified locality, suggests
that this species and arborealis may overlap in the lower Amazon but there is no in-
dication of intergradation of the 2 forms in the material at hand from Belem and up-
river. |

BIONOMICS. The immature stages of arborealis have been found in treeholes on-
ly. There are no published reports on the adult bionomics of arborealis but the rec-
ords of septemstriatus taken on human bait (Belem Virus Lab 1964) and during the
wet (Oct-Mar) and dry seasons (Causey and Santos 1949) undoubtedly refer to ar-
borealis in the present sense.

DISTRIBUTION (fig.2). Confined to low elevations in Guyana, Surinam, French
Guiana and Brazil (lower Amazon drainage). Material examined: 80 specimens; 8 6,
57 2, 6 pupae, 9 larvae; 7 individual rearings (3 larval, 4 incomplete).

BRAZIL. Para: Belem, July 1931, R.C. Shannon, 1 @ [USNM]; IAN Forest, 413 ?; IAN and
Utinga forests, 18 9; Utinga Forest, 107 ? (Belem Lab Report 1964:101,103,109,112). Curralinho,
1 Apr 1936, 1 9; Apr 1936, 1 ?; 26 July 1936, 1 3, 1 9; 1936, 1 L; 2 9 (K 325,341); Mar 1938, 5 9
[USNM]. Ilha de Marajo, 1 9 (Kumm and Novis 1938:507). Santarem, Belterra, Aug 1938, 1 9;
Oct 1938, 1 9 [USNM].

FRENCH GUIANA. Baduel (Floch and Abonnenc 1944:4; 1947:10). Cayenne, Apr 1943, 1 9
[USNM] ; June 1944, E. Abonnenc, 32 9. Cabassou, 4 Feb 1965, T.H.G. Aitken, R. Martinez and
A. Guerra (FG 46), 1 ¢ [UCLA]. Iracoubo (Fauran 1961:29). Placer Isnard, 17 June 1940, E.
Abonnenc, 2 9 [UCLA]. Rorota, 8 May 1941, 2 9, 3 lp (577,578) [USNM].

Berlin: Subgenus Howardina 25

GUYANA. Mazaruni, 25 June 1936, W.H.W. Komp, 2 2? [USNM]; 6 July 1936, Leslie, 1 6, 2 9
(KO 208A-7); 6 July, 1 6 (KO 208A-8); 1 6 (KO 208A-10) [UCLA].

SURINAM. Aloesoebanja, 17 Jan 1956, E. van der Kuyp, 1 L. Dam, 1 Ip, “‘cotype’”? [USNM,
No.22712]; 1 lp? (39-09, “‘cotype’’), 1 lp (39-01); 1 ¢ (38-98, “‘cotype”) [UCLA].

4. Aedes (Howardina) aureolineatus Berlin, n.sp.
Figs.2,8,58

. TYPES: Holotype ¢, Piraja (Bahia), Brazil, May 1931, Servico Febre Amarela, M.E.S., Brasil
[USNM]. Allotype ¢ (660121-10) with genitalia slide, same data as holotype [USNM]. Paratypes:
2 2, same data as holotype [USNM, UCLA].

FEMALE. Wing: 3.6 mm. Proboscis: 2.7 mm. Forefemur: 2.32 mm. Abdomen:
about 3.33 mm. Head: Median line of golden scales very narrow; these scales broader
than in fulvithorax and curved; submedian dark patch very large, extending to occi-
put; a small elongate lateral dark line at level of middle of apn. Palpus about 0.19 of
proboscis. Torus light yellowish brown, darkened on mesal surface. Thorax (fig.8):
Mesonotal integument brown; scaling predominantly dark bronzy; a conspicuous
acrostichal golden line from anterior promontory, bifurcating on prescutellar space
and reaching scutellum; a golden outer dorsocentral line arising caudad of scutal an-
gle, broad and curved outward anteriorly, narrow posteriorly and reaching scutellum;
lateral margin golden scaled, broadly to level of middle of ppn, narrowly to scutal an-
gle, broadly caudad of scutal angle, separated by dark scales from a very large golden
patch in and above supraalars; a few scattered golden scales laterad of acrostichal line
especially in front. Scutellum with a conspicuous median line of narrow golden
scales. Paratergite as in fulvithorax. Scales of apn numerous, light golden, moderately
broad and curved; ppn patch extensive, light golden, scales short and curved; remain-
ing pleural scale patches silvery to very light golden. Abdomen: Median basal tergal
light patches silvery on VI,VII, light brown on V, usually absent on other segments.

MALE. Wing: 3.06 mm. Proboscis: 3.1 mm. Forefemur: 2.16 mm. Allotype lack-
ing proboscis, 1 antenna, palpi and all legs on right side; description from Salvador
specimen. Essentially similar to female except that scattered golden scales more nu-
merous and lateral marginal light line wider on mesonotum. Palpus subequal to pro-
boscis in length.

MALE GENITALIA (fig.58). As figured; diagnostic characters as in the key. Es-
sentially as in arborealis except that lobe of tergite IX more prominent and several
setae of the claspette distinctly thickened.

IMMATURE STAGES. Unknown.

SYSTEMATICS. The adults of aureolineatus are strikingly different from all other
members of the group except sp. 5 (Iquitos form) in the reduction of the golden
scaling on the central anterior part of the mesonotum to a narrow acrostichal line.
This form appears to be a distinct species probably related to arborealis, but this can-
not be definitely determined until the immature stages are found.

In the small series of aureolineatus very little individual variation is evident. The
Specimens from Para and Rio de Janeiro are not significantly different from the type
material from Bahia.

BIONOMICS. Nothing is known of the breeding sites and adult bionomics of au-
reolineatus but it is likely that they are similar to other members of this group.

DISTRIBUTION (fig.2). Confined to low elevations in Brazil along the coast from
Para to Rio de Janeiro.

26 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

BRAZIL. Bahia: Piraja, May 1933, 1 d, 3 9, type series [USNM, UCLA]. Salvador, 27 Apr
1936, E. Silva, 1 6 [USNM]. Para: Apr 1930, N.C. Davis, 1 ° [USNM] . Rio de Janeiro: Apr 1933,
1 9 [USNM].

5. Aedes (Howardina) sp., Iquitos form

FEMALE. Wing: 3.33 mm. Proboscis: 2.71 mm. Forefemur: 2.34 mm. Abdomen:
broken off. Extremely similar to aureolineatus and differing apparently chiefly in the
following. Mesonotal ornamentation darker; acrostichal golden line narrower; dorso-
central golden line represented only by a rounded spot at about 0.5 of mesonotum;
fewer scattered golden scales; apn scales more numerous and broader, more distinctly
silvery; ppn scale patch more extensive, with some broad scales below.

MALE and IMMATURE STAGES. Unknown.

DISCUSSION. The single specimen from Iquitos, Peru, Mar-Apr 1931, R.C. Shan-
non [USNM] lacks the abdomen. It may represent a distinct species related to arbor-
ealis and aureolineatus or a local population of one of these. This form shows the
greatest reduction in the golden scaling on the anterior part of the mesonotum in the
Fulvithorax Group.

WALKERI SECTION

FEMALES. Usually small to medium sized species, wing 2.1-3.5 mm. Head: Sub-
median dark patch broad; lateral patch smaller, bordered posteriorly by light scales.
Erect scales varied. Torus brown to dark brown, with a few small black scales mesal-
ly. Palpus with or without apical light scales. Thorax: Mesonotum usually with dis-
tinct linear light pattern (except in some grabhami), light scales all golden or combi-
nation of golden and silvery. Acrostichal line absent but a small light patch present
at level of anterior promontory; inner and outer dorsocentral lines incomplete, inner
not reaching scutellum, outer usually not reaching anterior margin; lateral marginal
line broken at scutal angle, usually narrow (very broad in Walkeri Group); median
prescutellar line present. Paratergite bare. Acrostichal bristles usually absent on disc,
present in Busckii Group. Pleural scales silvery (fig.17); ppn and psp without scales;
ssp scale patch present or absent; mep with a small, restricted patch at upper half be-
low bristles only. Legs: Midfemur and hindfemur without distinct distal light spot on
anterior surface. Foretibia sometimes with light scales distally. Tarsal light markings
not restricted to segment 1. Wing: Veins entirely dark scaled. Abdomen: Sternites H-
IV without dark scales.

MALES. Larger claws of foreleg and midleg with both submedian and basal teeth
present (in Busckii Group basal tooth absent).

MALE GENITALIA. Lobes of tergite IX indistinct, bearing a few strong setae.
Sidepieces separated from each other at base by a distance greater than diameter of
aedeagus at that level; tergomesal and sternomesal margins without any indication of
lobes. Claspette represented by a small roughly triangular or fingerlike lobe bearing
1, rarely 2, apical setae; not connected to its mate.

PUPAE. Cephalothorax: Hair 5-C usually shorter than distance between its base
and that of trumpet; either subequal to or longer than hair 4-C; if longer, varied in
branching. Abdomen: Hair 5-IV,V usually single (double or triple in Aurites and
Busckii Groups), terminating before or extending beyond apex of tergite following;

Berlin: Subgenus Howardina | 8 |

2-IV-VI usually distad of 0.75 of its tergite; 9-III-VI usually caudolaterad of hair 6
(except in walkeri); 9-VII moderately to strongly developed, either close to or re-
moved considerable distance cephalad of hair 6.

LARVAE. Stellate hairs developed or not. Antenna: Short; with or without scat-
tered spicules; hair 1-A single. Thorax: Integument glabrous. Tubercles of hairs 5-7-P
distinctly joined; hair 9-P simple, usually single or double; hair 13-P absent except in
Ioliota Group. Abdomen: Integument glabrous except in walkeri and members of
Ioliota Group. Hairs 2,5,9-VII usually stellate; hair 8-VII cephalad or cephalodorsad
of 9-VII. Segment VIIT: Hairs 1,5-VIII usually strong, with a few branches but not
stellate. Comb scales usually in a single row (up to 2-4 irregular rows in Busckii
Group); individual scale usually blunt apically. Siphon: Index 2.0-8.5. Integument
smooth, imbricate or spiculate. Acus distinct except in Walkeri and Aurites Groups.
Pecten teeth with or without basal denticles. Anal Segment: Saddle with distinct
marginal spicules. Hair 1-X short to moderate, with a few branches; hair 3-X single,
2-X multiple. Ventral brush with 5 or 6 pairs of hairs arising from a boss with or
without distinct grid bars.

DISCUSSION. The Walkeri Section is the most heterogeneous of the 3 sections in
the subgenus, particularly in the immature stages. However, it is probably a mono-
phyletic lineage characterized by 3 unique features of adult ornamentation: parater-
gite and ppn both without scales and prescutellar area with a median light-scaled line.

Four rather distinct phylads within the section are suggested by differences in the
mesonotal pattern of the adults: (1) Aurites Group, with a rather broad linear golden
pattern very reminiscent of the Sexlineatus Section, (2) Walkeri Group, with a very
broad continuous lateral silvery patch and apparently relatively broad golden dorso-
central lines, (3) the very similar Ioliota and Albonotatus Groups, with a very narrow
silvery lateral line, very faint golden dorsocentral lines and a silvery prescutellar line,
and (4) Busckii Group, intermediate between the Walkeri Group and the Ioliota-
Albonotatus Groups in the width of the lateral silvery line which is frequently broken.
The Busckii Group is unique in the subgenus in the presence of acrostichal bristles on
the disc of the mesonotum.

The immature stages are much more diverse than in the other sections, each of the
above-mentioned phylads possessing a type quite as distinct as those characterizing
the Sexlineatus and Fulvithorax Sections. The Ioliota Group is also clearly separated
in the larval and pupal stages from the Albonotatus Group. The breeding sites in the
Walkeri Section encompass the entire range used by the subgenus, extending to habi-
tats not utilized by the other sections.

The Walkeri Section is almost entirely confined to islands in the Caribbean area.
Only 2 small encroachments on present continental areas are known: (1) in the Yu-
catan Peninsula by the Ioliota Group and (2) in northern Venezuela and Trinidad by
the Ioliota Group and northern Venezuela only by the Walkeri Group. The groups
show primarily allopatric complementary distributions. The only exceptions known
are in (1) the Walkeri and Aurites Groups in Jamaica, (2) possibly Albonotatus and
Busckii Groups in Guadeloupe, and (3) Ioliota and Walkeri Groups in northern Vene-
zuela. Except in the Aurites Group, members of a group always have allopatric dis-
tributions (fig.3).

There are 2 surprising gaps in the distribution of the Walkeri Section. Although in-
tensive collections have been made in Puerto Rico, Virgin Islands and Leeward Is-
lands, no species of Howardina has yet been recorded from this area. Nor has the
subgenus been recorded from Cuba. However, it is possible that the record of Haema-
gogus sp. from Cuba by Montchadsky and Garcia (1966:46-47) is based on a mis-

28 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

identification of an undescribed species of Howardina.

The diversity of types in the Walkeri Section is undoubtedly a reflection of typical
insular speciation through isolation of small populations. As indicated above, it ap-
pears that only one original stock was probably involved, although there is a sugges-
tion of a possible recent intrusion of the Sexlineatus Section in the Aurites Group
(see). However, the present pattern could not have been attained through a simple
splitting of the original stock among the island masses as they exist today. The reten-
tion of such primitive features as acrostichal bristles in the adults of busckii, hair 13-
P in the larvae of the Ioliota Group, 6 pairs of hairs in the ventral brush in all species
except busckii and a patch of comb scales in the larvae of some populations of
busckii, is indicative of very early separation of the Walkeri Section from the other
sections rather than its recent derivation from one of them. That the present pattern
represents speciation and dispersal on island masses no longer in existence in the
Caribbean area is strongly supported by the relict distributions of the Walkeri and
Ioliota Groups.

KEYS TO GROUPS AND SPECIES

ADULTS
3 All mesonotal leht scaling golden (Aurntes Group)... 2°. we
At least lateral marginal lime silvery-or white <0 2. 2 7) cosue Teeio nw a

Aurites Group

2(1). Pleuron with a large silvery patch continuous from lower pra across posterior

stp toitslower margin. . . - oc 6) aurives
Pleuron with 3 small discrete, separate silvery parches. one each on lower
Pra, Upoct. se and {ower 870 (6 Oe A ee

3(2). Antealar area (above paratergite) with golden scales, rarely 1-3 bronzy scales
present; mesonotal light lines usually distinct and broad, sometimes almost
completely merging in anterior two-thirds of mesonotum . .9. grabhami

Antealar area with numerous dark bronzy scales above narrow posterior lat-
eral mesonotal golden line; mesonotal light lines narrow, discrete, never
Mere ie ee Oe IO imeeamalis, 1 ic siene

4(1). Lateral marginal line broadened to form a conspicuous lateral silvery patch;
ssp without scales (Walkeri Group) .
Lateral marginal line very narrow, not extending o on fossa: - ssp with patch of

ls yes atin cen ee
Walkeri Group

5(4). Inner dorsocentral line continuous, moderately broad. . >. . . 6. walkeri

Inner dorsocentral line dotted, only faintly indicated . . . . . 7. argyrites
Busckii Group

6(4). Median prescutellar line golden, faint; palpus entirely dark in both Sexes;
acrostichal bristles presenton disc @ 2) 0 a eee a 12 busekil

7(6).

8(7).

(7).

oul

S(2).

4(3).

=),

Berlin: Subgenus Howardina 2

Median prescutellar line silvery, conspicuous; silvery scales on apex of female
palpus and base of some segments of male palpus; acrostichal bristles ab-

Sent OHCne a ee ee
Hindtarsal segment 5 with ae scales (Ioliota Group) eg ee ae
Hindtarsal segment 5 with black scales only (Albonotatus Group). re

Ioliota Group

Hindtarsal segment 5 entirely white-scaled . . . tues 6G 2. toloda
Hindtarsal segment 5 with some apical and posterior dark scales .
ws comnclensis
Albonotatus Group

Hindtarsal segment 3 white-scaled in barely more than basal 0.5 . ;
Mee ee a) eg
Hindtarsal segment 3 white-scaled in at least basal0.7. . . 14. albonotatus
MALE GENITALIA
(7. argyrites not definitely known)

Tergal surface of sidepiece with very dense vestiture of long bristles, some as

long as’sidepiece . . >. 5 a. LD. Daehamensis
Tergal surface of sidepiece with sparse vestiture of short or moderate bristles,
none more thay.0.5’of sidepiece length 27 ee? Gk ee

Clasper very a including spiniform about as peice as sidepiece .
14. albonotatus
Clasper short or r moderate, including spiniform not more than 0.75 of side-
eieceleneih.. . -. Caete cromp ik Ged (OILS peak ig eine ee

Claspette with an apical fingerlike cylindrical process more than twice as long
as wide and bearing apical bristles; body of claspette with short subapical
Drisiies: 2 ae. 4

Claspette without distinct apical fingerlike cylindrical process, apical bristle
borne on a short broad conical process; body of claspette with several
strong subapical bristles borne on similar but smaller tubercles . . . .5

Apex of aedeagus deeply emarginate in dorsal aspect producing lateral lobes
and/or distal half of aedeagus distinctly broader than proximal half; basal
tergal bristles of sidepiece strong, subequal in length and development to
tergomesal marginal bristles. . . . pu Soe EO ser

Aedeagus not emarginate apically in dorsal aspect and roughly of uniform
width; basal tergal bristles of sidepiece weak, much shorter and weaker
than tergomesal marginal bristles

6. aurites, 9. grabhami; | 0. inaequalis; : a stenei

Clasper more than 0.5 of sidepiece length, without distinct subapical setae .
6. walkeri
Clasper less than 0. 5 of sidepiece length. with distinct subapical eetae <0

30

6(5).

2(1).

3(2).

4(3).

a¢3).

6(2).

7(6).

8(6).

9(8).

Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Clasper gradually tapered todistalO.75 ..... . .13. cozumelensis
Clasper swollen on base, abruptly tapered at about middle So wh pedoge
PUPAE

(7. argyrites unknown)

Walkeri Group

Paddle deeply emarsinate athar!] .°. . 6. . . . 6. walkeri
Paddle usually smoothly rounded on apex, at ctost shallowly emarginate at

PR a a ei as ee ee ds ee ee
Hair 10-C usually at least 10-branched (Aurites ee Sy se ak a peace
Hair 10-C usually less than 5-branched. . .. . ee en eek

Aurites Group

Pigmentation very light and uniform; metanotum pene scape pigmen-
ted ore.
Pigmentation moderate and t uneven; metanotum with light < area in the middle

er ro a ee
Hair 2-¥ idistinctly laterad Of Nair 3. 4 ea ee ses hae Ge oe © UTES
Flaw 2-Vi miesad or at level of Neir3. . 6 Om es BR Bona

Median tergal darkening of abdominal segment III indistinct.
ie SC, inaequalis
Median tergal darkénine of abdominal segment i sharply marked |

11. stenei
Hair 5 at least 2.0 of hair 4 (loliota Group). . . ae aoe
Hair 5-C subequal to hair 4 (Albonotatus and Busckii Groups). Sree

Ioliota Group

Hair 1-II strongly dendritic and with many branches; paddle ovoid, length

about 1.5 greatest width . . i G2. toon
Hair 1-II with 7-12 simple branches: ‘paddle rounded, length about 1.2 great-
OS Wiltd ue Beak Begs err Gay ah ane, hy eyes

Albonotatus and Busckii Groups

Hairs 7,10-C both Single; paddle elongate, length more than 1.5 greatest

width . . 14. albonotatus
Hairs 7 10-C usually at least double: paddle length moderate, usually less
than 1. 3 greatest Widths so ucan he ee

Pinna very long, more than 0.33 of total trumpet length; paddle slightly
emarginate at hair 1; hair 6-III],IV strong, usually at least double .

. . L535. bahamensis

Pinna moderate, about 0. a of total trumpet length: paddle evenly rounded

on apex; hair 6-III AV Weak Wives Pe ea ee, Base

2(1).

erg

4(2).

5(4).

6(5).

At).

8(7).

9(8).

Berlin: Subgenus Howardina 31

LARVAE

(7. argyrites unknown)

Walkeri Group
Abdominal integument densely covered with large dark-branched spicules .
oc Bia Gm Me Te ee el ey ee: 6. walkeri
Abdominal injeeument Slanraus at A a a ee es ee ee
Hair 13-P present (loliota ack ae WT ee RESET OO NO COE ini Sarna NACE RE
Hair 13-P' absent ... 3 fMea Gk eet lea ige Ugly eas ees ae eae

Ioliota Group

Siphon slightly swollen in middle, index about 2.5; comb scales blunt or

rounded apically... . ped £2 Aptliota

Siphon slightly evenly tapered from base, index. about a OQ; comb scales

Ponte APIO ee I I aa ions
Busckii Group

Ventral brush with 5 pairs of hairs (comb scales sometimes in 2 to 4 rows;

pecten teeth sometimes extending well beyond hair 1-S). . . 16. busckii

entrar Dish) Will G paws OL ais” Ce ee eee

Head hair 6-C usually single or double; pecten teeth short and with basal den-
ticles; siphon index 3.0 or less (Albonotatus Group) . . .6
Head hair 6-C multiple, usually more than 5-branched; pecten teeth long,
slender and without basal denticles; siphon index more than 4.0 (Aurites
eC) petunia ee ee at

Albonotatus Group

Anal gills subequal; siphonal hair 1-S adjacent to last pecten tooth .
. . . 14. albonotatus
Anal ills markedly unequal siphonal hair 1. -S well beyond last pecten tooth
ool Eh ae ERR cs £5, bahamensis

Aurites Group

Hair 2-VII large, stellate, 4-7-branched. . ...... . .. .. Jd. stenei
PROT ee i IR OR ie iar dyna Wie ecw aga a age eee ee
Siphon index less than 5.5; anal gills markedly unequal, the small distinctly
shorter than saddle... |... . . . 10. inaequalis
Siphon index 6.0 or more; anal gills subequal and distinctly longer than sad-
1 Nill idle Coe ihdib inline ree Rue) a Bi) Lene

Head hair 5-C usually single; 7-C very short, less than 0.25 of antenna, and
usually triple . . ay? grabhami
Head hair 5-C a at least triple; +e about 0. 5 of antenna and usually 6-
DYANENOG 464. eae gc ee Na ie ae ee ee ee ae

32 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Walkeri Group

FEMALES. Head: Narrow decumbent scales white. Palpus with apical silvery
scales (broken in argyrites). Thorax: Inner dorsocentral line golden, continuous or
dotted; median prescutellar line broad, white or predominantly white; outer dorso-
central line faint, narrow, golden; lateral marginal line broadened to form a conspic-
uous lateral silvery patch from anterior angle to wing base. Acrostichal bristles not
developed on disc: Scutellum predominantly dark scaled, light scales restricted to
median lobe. Ssp without scales; upper stp patch separate, large. Legs: Foretarsus
completely dark; midtarsus with incomplete or complete basal white ring on segment
1; hindtarsus with moderate basal white rings on segments 1-3 in walkeri, at least on
1-2 in argyrites (rest broken). Abdomen: Basomedian tergal light markings creamy,
present on proximal segments.

MALE (walkeri only). Palpus with basomedian. white scales on segments 4,5. Larg-
er claw of foreleg and midleg with both submedian and basal teeth; smaller claw with
basal tooth only.

MALE GENITALIA (walkeri only). Caspette with short broad conical tubercle
bearing 1 apical bristle. Clasper about 0.6 of sidepiece; minute spicules basally; sub-
apical hairs absent. Aedeagus subquadrate, shorter than proctiger, without apical
~ lobes.

PUPA (walkeri only). Cephalothorax: Hair 5-C usually at least triple, moderately
long, subequal to distance from its alveolus to trumpet base and at least 2.0 as long
as hair 4; hair 7-C usually 3-branched; hair 10-C with 2-7 branches. Abdomen: All
large hairs except 1-II,IJI, 9-VII,VHI usually single or double; hair 2-II laterad of or
in line with hair 3; hair 5-[V,V usually extending to apex of tergite following; hair 9-
II-VI usually at level or slightly cephalad of hair 6; 9-VII subequal to hair 6-VIJ; hair
4-VIII single. Paddle: Short; apex distinctly emarginate; marginal spicules distinct.

LARVA (walkeri only). Head: Hair 1-C always branched, usually triple; 4-C un-
usually long, multiple; 5-C single, long and stout; 6-C usually 4-branched; hairs 11,
12,14-C all stellate, 11-C with unequal branches; 15-C double, long, exceeding apex
of mental plate. Antenna: Shaft with scattered, short, sharp spicules; hair 1-A single.
Thorax: Hair 13-P absent. Hairs 0,1,3,4,8,14-P, 1,13,14-M, 1,3,4,5,8,13-T all strong-
ly stellate, with numerous branches, very darkly pigmented; hairs 1-M,T far mesad of
~ hairs 2,3. Abdomen: Integument with numerous very dark, large branched spicules;
hairs 2 and 9 removed cephalad of hair 8. Hairs 1,2,5,9,13-I-VI, 11-I, 7-II-VI all
strongly stellate as on thorax; hair 8-VII cephalodorsad of 9-VII. Segment VUIIT:
Comb scales in a single row; individual scale long, pointed and fringed to apex. Si-
phon: Index less than 3.0; integument with rows of short spicules; acus absent, no
indication of basal indentation. Pecten extending almost to apex; teeth long, proxi-
mal ones simple, distal branched apically. Anal Segment: Saddle body with numer-
ous strong spicules; caudal margin with long fringed spines; hair 1-X multiple, near
lower end of caudal margin. Ventral brush with 6 pairs of hairs on very strongly scle-
rotized boss without distinct grid bars.

DISCUSSION. The diagnosis of the group is based almost entirely on walkeri, the
type species of the subgenus and a common species endemic to Jamaica. A. argyrites,
known only by the imperfect female holotype from northern Venezuela and possi-
bly a male genitalia slide, is extremely similar to walkeri in adult ornamentation and
is undoubtedly a member of this group.

The Walkeri Group possesses many features unique in the section, particularly in
the immature stages which are readily recognized from all other species in the sub-

Berlin: Subgenus Howardina 33

genus. However, the mesonotal pattern of the adults suggests a greater affinity with
the Aurites Group than with any other group. As indicated under the latter, this is
interpreted here as an indication of introgression with the Sexlineatus Section.

The widely disjunct distribution of the 2 members of the Walkeri Group parallels
that of the 2 members of the Ioliota Group and is similarly interpreted as a relict pat-
tern of a group originally with a wide distribution in the central Caribbean area (fig.
B).

The group may be primarily a breeder in bromeliads as is the case with walkeri but
since the latter sometimes uses treeholes it is possible that these may have been the
original breeding sites, perhaps still retained by argyrites.

6. Aedes (Howardina) walkeri (Theobald)
Figs.3,7,8,16-18

1901.Culex walkeri Theobald, 1901:424. TYPE: Holotype ? (45/110), Jamaica, date and collector
not specified [BM].

Aedes (Howardina) walkeri of Bonne and Bonne-Wepster (1925:368); Dyar (1928:235); Edwards
(1932:156); Lane (1939:101; 1953:725-726); Horsfall (1955:471); Stone, Knight and Starcke
(1959:174); Belkin, Schick and Heinemann (1965:31); Forattini (1965S: 399 ,400); Porter (1967:
38).

Aedes walkeri of Dyar and Knab (1906b:189,192); Howard, Dyar and Knab (1912:70; 1917:849-
852); Picado (1913:325); Dyar (1918:73,80); Johnson (1919:424); Gowdey (1926:74); Thomp-
son (1947:78); Hill and Hill (1948:46).

Howardina walkeri of Theobald (1903a:287-289); Dyar (1905a:49; 1905b:23; 1905c:27-28); Theo-
bald and Grabham (1905:20-22); Theobald (1905:9; 1907:215-216; 1910:220); Coquillett
(1906b:26); Grabham (1906:170-171); Surcouf and Gonzalez-Rincones (1911:139-140).

Culex walkeri of Blanchard (1905:416).

Haemagogus walkeri of Dyar and Knab (1906a:166).

FEMALE (fig.16). Wing: 2.62 mm. Proboscis: 2.1 mm. Forefemur: 1.75 mm. Ab-
domen: about 2.45 mm. As described for group, with the following apparently diag-
nostic features. Head: Median stripe on vertex broad, silvery; light scales on sides and
venter creamy. Thorax (figs.8,16): Scales on anterior promontory golden. Inner dor-
socentral line distinct, continuous and moderately broad; median prescutellar line
usually silvery, sometimes with a few golden scales, rarely predominantly golden;
outer dorsocentral line golden; lateral marginal line very broad, silvery. Antealar area
above paratergite with silvery scales. Silvery scales on median scutellar lobe narrow;
lateral lobes with narrow dark scales. Pleural scales as in the group. Legs: Forefemur
and midfemur predominantly dark anteriorly, a few basal scales creamy, proximal
0.5 of posterior surface broadly creamy, extending distally as a light line, rest dark;
hindfemur mostly creamy except for lateral distal dark scales. Midtarsal segment 1
with only a few white scales at base. Haltere: Stem pale, knob mostly dark, fringe
scales creamy white. Abdomen: Basomedian tergal creamy patches distinct on {I-VI
a few pale scales on II. Sternites creamy on II-IV, rest of segments with basal creamy
and apical dark scales.

MALE (fig.16). Wing: 2.28 mm. Proboscis: 1.93 mm. Forefemur: 1.04 mm. Essen-
tially similar to female in color.

MALE GENITALIA (fig.17). As figured; diagnostic characters as in the key. Ninth
tergite lobe indistinct, with 3,4 strong setae. Sidepiece about 2.5 times median width.

34 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Clasper with minute spicules basally; spiniform about 0.2 of clasper.

PUPA (fig.17). Abdomen: 2.7 mm. Trumpet: 0.38 mm; index about 4.5-5.0. Pad-
dle: 0.46 mm. Chaetotaxy as figured; hairs moderately to darkly pigmented. Diagnos-
tic characters as in the key. Readily separated from all other species of the subgenus
by the presence of minute denticles on abdomen and by the distinctly emarginate
paddle. Cephalothorax: Mesonotum moderately pigmented between trumpets, slight-
ly deeper laterally; trumpets contrasting. Hair 1(1), 2(2, 2-3), 3(1, 1-2), 4(3, 2-5),
5(5, 3-6), 6(1, 1-2), 7(3, 2-4), 8(3, 1-5), 9(1), 1005, 2-7), 1101, 1-2), 12(3, 2-5). Ab-
domen: Tergite I moderately to strongly pigmented except for usual membranous
areas; remaining segments moderately to strongly pigmented, progressively lighter
caudad; integument with minute denticles. Segment I: hair 1(8-12 primary branches,
each with many secondary branches), 2(1, 1-2), 3(1), 4(3, 2-5), 5(4, 3-6), 6(1, 1-2),
7(2, 1-3), 9(1, 1-2). Segment II: hair 0(1), 1(8,.5-17), 2(1), 31), 4(@2, 2-4), 5@, 1-3),
6(1, 1-2), 7(3, 1-4), 81, present in some), 9(1). Segment III: hair 0(1), 1(4, 2-9), 2€1,
1-2), 3(1), 4@, 1-3), 5(2, 1-3), 6(1), 7(2, 2-5), 8(3, 2-5), 9(1), 10(1, 1-2), 11(1), 14(1).
Segment IV: hair 0(1), 1(3, 2-4), 2(1, 1-3), 3(3, 2-6), 4@, 1-2), 5(2, 2-3), 6(1, 1-2),
7(2, 1-4), 8(2, 2-3), 9(1), 101, 1-3), 11€1), 14(1). Segment V: hair 0(1), 1(2, 1-3),
2(1, 1-3), 3(1, 1-3), 41, 1-5), 5@, 2-4), 6(1, 1-2), 7(2, 2-5), 8(2, 2-4), 9(1), 101, 1-
2), 111), 14(1). Segment VI: hair 0(1), 1(1, 1-3), 2(1, 1-3), 31, 1-2), 4(2, 1-4), 5(3,
2-4), 6(1, 1-2), 7(2, 1-3), 8(2, 2-3), 9(1), 10(1), 11€1), 14(1). Segment VII: hair 0(1),
1(1, 1-3), 21, 1-3), 3(2, 1-3), 4(1, 1-2), 5(@2, 1-4), 6(5, 3-6), 7(1, 1-2), 8(2, 2-6), 9(6,
6-9), 10(1), 11(1, 1-4), 14(1). Segment VIII: hair 0(1), 4(1), 9(12, 11-16), 14(1, 1-2),
Paddle: Uniformly lightly pigmented, midrib slightly darker; dorsal surface with fine
denticles, external margin serrated; apex deeply emarginate; hair 1-P slender. Male
genital lobe extending to 0.5 of paddle, female genital lobe to 0.41.

LARVA (fig.18). Head: 0.7 mm. Siphon: 0.43 mm. Anal Saddle: 0.22 mm. Chae-
totaxy as figured; hairs darkly pigmented. Diagnostic characters as in the key. Readi-
ly differentiated from all other species of the subgenus by: (1) stellate hairs with nu-
merous branches, (2) dendritic spicules on abdomen and (3) pecten teeth extending
almost to apex of siphon. Head: Width subequal to length; integument smooth and
strongly pigmented. Mental plate with 12 distinct teeth on each side of strong medi-
an tooth, often a smaller detached tooth at base. Hair 0(1), 1(3, 2-3), 3(1), 4(6, 5-8),
5(1), 6(4, 2-5), 7(4, 3-5), 8(1), 9(2), 10(€1), 11(14, 9-17; stellate, branches of 3
lengths), 12(10, 8-17), 13(1), 14(15, 11-24; stellate), 15(2, 1-3). Antenna: Length
about 0.3 of head; uniformly strongly pigmented; with scattered spicules; all hairs
usually single; hair 1-A about 0.62 from base. Thorax: Integument glabrous; in life
thorax appears bluish green. Apex of long hairs attenuate; stellate hairs with numer-
ous branches; long and stellate hairs weakly barbed. Prothorax: hair 0(32, 30-46),
1(32, 27-38), 2(1), 3(26, 24-36), 4(36, 30-42), 5(4, 3-5), 6(1, 1-2), 7(3, 2-4), 8(36,
30-38), 9(2, 1-4), 10(1), 11(4, 3-5), 12(1), 14(48, 44-54). Mesothorax: hair 1(36, 25-
54), 2(1), 31), 4C1), 5(1), 6(8, 5-8), 7(1), 8(8, 7-10), 9(10, 7-14), 10(1), 11(3, 2-4),
12(1), 13(22, 17-27), 14(23, 19-29). Metathorax: hair 1(54, 48-63), 2(1), 3(22, 19-
24), 4(28, 28-34), 5(25, 22-33), 6(1), 7(8, 7-10), 8(19, 16-23), 9(8, 6-8), 10(1, 1-2),
11(3, 2-4), 12(1), 13(42, 38-43). Abdomen: Integument densely packed with large
dendritically branched spicules, more sparsely on segment I; stellate hairs as on tho-
rax. Segment I: hair 1(53, 51-71), 2(40, 40-53), 3(1), 4(2, 1-2), 5(29, 28-41), 6(3,
3-4), 71, 1-2), 9(14, 9-15), 10(1), 11(26, 24-30), 13(35, 29-38). Segment II: hair
O(1), 1(60, 52-65), 2(46, 36-46), 3(1), 4(2, 1-3), 5(32, 25-36), 6(4, 3-4), 7(18, 14-
22), 8(1), 9(20, 18-22), 10(1), 11(1), 12(1), 13(36, 33-44). Segment III: hair 0(1),
1(64, 56-66), 2(41, 35-45), 3(1), 4(1), 5(34, 30-40), 6(3, 2-3), 7(20, 20-24), 8(1),

Berlin: Subgenus Howardina 35

9(21, 21-26), 10(1), 11(1), 12(1), 13(36, 32-44), 14(1). Segment IV: hair 0(1), 1(60,
54-62), 2(42, 36-46), 3(1), 4(1, 1-2), 5(37, 30-42), 6(2, 2-3), 7(20, 18-22), 8(1),
9(24, 22-27), 10(1), 11(1), 12(1), 13(40, 25-41), 14(1). Segment V: hair 0(1), 1(60,
47-61), 2(37, 37-42), 3(1), 4(3, 1-3), 5(36, 30-41), 6(@2, 2-3), 7(20, 18-25), 8(1),
9(26, 25-31), 10(1), 11(1), 12(1), 13(36, 30-42), 14(1, 1-2). Segment VI: hair 0(1),
1(56, 44-60), 2(41, 36-41), 3(1), 4(2, 1-2), 5(30, 30-42), 6(2, 1-2), 7(22, 18-24),
8(1), 9(28, 25-29), 10(1), 111), 1201), 1334, 28-38), 14(1). Segment VII: hair 0(1),
1(50, 42-50), 2(36, 30-40), 3(1), 4(1), 5(28, 28-39), 6(20, 16-26), 7(1), 8(24, 22-
28), 9(26, 26-32), 10(1), 1101), 12(1), 13(22, 20-28), 14(2). Segment VIIT: Comb
scales in a single row of 12(10-14); individual scales long, pointed apically, fringed to
the tip. Hair 0(1), 1(48, 46-54), 2(1), 3(8, 8-10), 4(1), 5(28, 26-31), 14@, 1-2). Si-
phon: Index about 2.8-2.9; darkly pigmented; integument uniformly pilose. Pecten
extending beyond hair 1-S, almost to apex; 20(16, 22) teeth, progressively longer dis-
tad, individual teeth long, without basal denticles, distal teeth broad, with apical den- ©
ticles. Hair 1-S about 0.67 from base. Hair 1(3, 2-4), 2(1), 3-5 not visible, 6(1), 7(1),
8(1), 9(1). Anal Segment: Saddle strongly pigmented, with darker basal ring; integu-
mentary spicules progressively larger distally; caudal margin with long spines fringed
laterally. Gills narrow, moderately long, ventral pair slightly shorter than dorsal. Hair
1(8, 6-10; strong, at caudolateral angle), 2(12, 10-14), 3(1), 4a(7, 6-8), 4b(7, 6-7),
4c(6, 5-8), 4d(6, 6-7), 4e(6, 6-7), 4f(S, 4-7).

EGG. Length about 0.62 mm, width about 0.14 mm; exochorion transparent and
endochorion dark and with polygonal sculpturing.

SYSTEMATICS. A. walkeri, the type species of Howardina, is among the most
clearly marked species in the subgenus. In the adults, only the closely related argyri-
tes, known by a single female from Venezuela, could be confused with walkeri. The
immature stages are also very strikingly differentiated but it would not be surprising
if those of argyrites, when found, were to conform to the same general type in view
of the resemblance of the females.

Little variation is apparent in the extensive material of walkeri, either in the adults
or the immature stages. Only the median prescutellar line exhibits some noticeable
variation, ranging from entirely silvery in the majority of specimens to predominant-
ly golden in a few. The dorsocentral lines are always clearly marked and never dotted
- aS in argyrites.

BIONOMICS. The breeding sites of walkeri are usually bromeliad axils but a few
collections of immatures have been made in treeholes (JA 336,390), broken and cut
bamboo (JA 711,778), heliconia axils (JA 748) and a crabhole (JA 369). Both terres-
trial and arboreal bromeliads are used by walkeri. This species has been found associ-
ated with all members of the Aurites Group and appears to have a much wider alti-
tudinal and geographical range than any other species of Howardina on Jamaica. The
larvae of walkeri are readily recognized in the field by their very dark and “hairy”
bodies.

Females of walkeri commonly bite humans under the shade of bromeliad laden
trees. I found a maximum peak of activity in mid morning (0900) and a lesser one in
late afternoon (1700) as shown in fig.7. However, Hill and Hill (1948:46) state that
walkeri prefers to feed at dusk in the open, and bites at any time of the day in forest
shade.

DISTRIBUTION (fig.3). Endemic to the island of Jamaica where it is widespread
at elevations from 100 ft to 4500 ft. Material examined: 2964 specimens; 244 4,
1139 2, 699 pupae, 882 larvae; 516 individual rearings (237 pupal, 150 larval, 129
incomplete).

36 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

JAMAICA. Clarendon: Honeycomb Rock, elev. 800 ft, 8 Aug 1965, W.A. Page (JA 237), 1 Ip
(237-27) [UCLA]. Mocho Road, 10 Sept 1965, J.N. Belkin, 1 9 (JA 353) [UCLA]. Slippery Fall,
elev. 800 ft, 8 Aug 1965, W.A. Page, 1 L (JA 239) [UCLA]. Hanover: Copse, elev. 500 ft, 6 Nov
1966, O.G.W. Berlin and D.C. Watson (JA 668), 1 pd (668-101), 1 p? (668-100), 6 L, 21 [UCLA].
Lethe, elev. 500 ft, 6 Nov 1966, O.G.W. Berlin and D.C. Watson, 4 p, 1 L, 5 1(JA 655), 1 lp? (667-
10), 5 pd (667-100-104), 5 IP (667-11-15), 2 9,2 P, 4 p, 34 L, 41(JA 667) [UCLA]. Manchester:
Hope Farm, elev. 1000-1100 ft, 10 Sept 1965, W.A. Page, 1 p? (354-102), 1 IP (354-13), 55 9 (JA
356); 23 Oct 1966, O.G.W. Berlin and D.C. Watson, 31 2 (JA 634A), 2 Ipd (635-11,16), 2 Ip? (635-
10,12), 4 pd (635-100-103), 1 p? (635-104), 8 IP (635-13-15,17-21), 1 P, 21 (JA 635); 2 Nov
1966, O.G.W. Berlin, 12 2 (JA 663A), 779 2 (JA 684) [UCLA]. Mandeville, elev. 900 ft, 23 Oct
1966, O.G.W. Berlin and D.C. Watson (JA 639), 4 lp? (639-10,14,22,25), 8 pd (639-101,104,107-
111,114), 7 p? (639-100-103,105,106,112,113), 12 IP (639-11-13,15-21,23,24), 9 6, 4 9, 18 P,
22 p, 62 L, 24 1; 24 Oct 1966, O.G.W. Berlin (JA 640), 1 IP (640-10), 1 L [UCLA]. Melrose Hills,
elev. 900 ft, 8 Aug 1965, W.A. Page (JA 244), 1 lpd (244-43), 1 p? (244-41A), 1 IP (244-40), 7 P
[UCLA]. Porus, elev. 900-1000 ft, 8 Aug 1965, W.A. Page (JA 243), 2 Ipd (243-22,29), 3 lp?
(243-21 ,26,27), 1 pd (243-102), 2 p? (243-101,103), 2 IP (243-19,20), 1 P, 15 L; 23 Oct 1966,
O.G.W. Berlin and D.C. Watson, 1 IP (631-30), 2 Ip? (633-11,12), 1 p? (633-100), 5. IP (633-10,13-
16), 4 L (JA 633); 1 Nov 1966, O.G.W. Berlin (JA 657), 2 lpd (657-10,11), 1 pd (657-100), 2 p?
(657-101,102), 2 IP (657-13,14), 2 P, 2 p, 1 L, 7 1; 2 Nov 1966, O.G.W. Berlin, 1 lpd (660-10),
1 pd (660-100), 1 P (JA 660), 1 Ip? (664-12), 4 pd (664-100,103-105), 2 p? (664-101,102), 2 IP
(664-10,11), 1 P, 22 L(JA 664) [UCLA]. Williamsfield, elev. 1100 ft, 23 Oct 1966, O.G.W. Berlin
and D.C. Watson (JA 637), 4 Ipd (637-10,11,13,17), 3 lp? (637-12,23,27), 10 pé (637-100-105,
108-110,112), 3 p? (637-106,107,111), 11 IP (637-14-16,18-22,24-26), 3 p, 25 L, 1 1 [UCLA].
Portland: Ecclesdown, elev. 1000 ft, 22 Aug 1967, J.N. Belkin and W.A. Page (JA 748), 1 p? (748-
94), 1 p [UCLA]. Fairy Hill, elev. 100 ft, 13 Nov 1966, O.G.W. Berlin and D.C. Watson, 20 L, 21
(JA 693), 4 Ipd (694-11,13,14,17), 9 lp? (694-10,12,15,18-23), 21 pd (694-100-103 ,106-108,112-
117,119-121,123,125,128-131), 13 p? (694-101 ,104,105 ,109-111,118,122,124,126,132,133), 1.1P
(694-16), 4 6,1 9,2 P, 2 p, 34 L (JA 694), 19 L (JA 695) [UCLA]. Kemney-Sherwood Forest,
elev. 750 ft, 13 Aug 1967, J.N. Belkin and W.A. Page (JA 707), 3 lpd (707-10-12), 2 pd (707-100,
102), 1 IP (707-13), 1 P, 1 L; 13 Aug 1967, W.A. Page, 2 Ipd (708-11,12), 1 lp? (708-15), 1 d(JA
708), 1 Ipd (709-10), 1 pd (709-101), 1 p, 1 L(JA 709), 1 L(JA 710), 1 lpd (716-10), 3 pd (716-
100,101,104), 1 P, 4 L (JA 716) [UCLA]. Portland Point, 4 Aug 1942, C.B. Philip, 1 @ [USNM].
St. Andrew and Kingston: Castleton, elev. 600 ft, 27 Oct 1966, O.G.W. Berlin and D.C. Watson
(JA 645), 2 IP (645-10,11), 1 P, 5 L, 5 1 [UCLA]. Duhaney River, 6 Nov 1965, W.A. Page (JA
399), 1 IP (399-13) [UCLA]. Ferry Quarry, 13 Sept 1965, W.A. Page, 1 d (JA 369) [UCLA].
Hardwar Gap, elev. 3800 ft, 29 Oct 1966,0.G.W. Berlin (JA 647), 2 lpd (647-10,11), 1 lp? (647-
12), 1 pd (647-100), 1 p? (647-101), 1P, 1 p, 1 L, 11 [UCLA]. Hermitage Dam, 1934, 6 9? [USNM].
Hermitage Dam Road, 22 Aug 1965, W.A. Page (JA 282), 1 Ipd (282-36), 1 Ip? (282-33), 1 IP (282-
31); 13 Aug 1965, W.A. Page, 1 IP (294-44), 1 IP (304-21); 1 Sept 1965, W.A. Page, 1 Ipd (319-22),
1 Ip? (319-23), 1 pd (319-104), 31 L(JA 319), 1 lp? (322-23), 3 IP (322-24,31,32), 1 9 (JA 322),
3 L (JA 323), 2 p? (324-102,103), 6 L (JA 324); 7 Sept 1965, J.N. Belkin (JA 333), 1 lp? (333-
21), 1 pd (333-100), 2 p? (333-101 ,102), 1 IP (333-20), 2 L; 29 Oct 1965, W.A. Page, 1 9 (JA 390),
1 P (JA 391), 1 Ipd (392-12), 1 Ip? (392-11), 6 pd (892-102,104,106,109,111,113), 4 p? (392-
100,101,105,114), 1 IP (392-10), 2 3, 2 9,2 p, 16 L(JA 392), 1 lpd (393-12); 15 Aug 1967, J.N.
Belkin and W.A. Page, 1 Ip? (723-30), 1 pd (723-103), 1 p? (723-102), 2 L(JA 723), 2 Ipd (724-
12,14), 3 Ip? (724-10,11,15), 3 pd (724-102-104), 1 IP (724-13), 1 P, 2 p,3 L(JA 724), 2 Ipd
(725-15,16), 1 Ip? (725-14), 5 pd (725-90,100A,105A,106,108A), 3 p? (725-92,107,112), 3 3,
3 2,1 P,3 p (JA 725); 1 Sept 1967, W.A. Page and J.N. Belkin (JA 765), 1 IP (765-10), 2 L; 13
July 1967, W.A. Page (JA 843), 3 lpd (843-12,13,15), 3 Ip? (843-10,11,14), 1 pd (843-101), 2 p?
(843-100,102); 25 July 1967, W.A. Page, 1 L (JA 854), 1 Ipd (855-11), 2 lp? (855-1012), 3 IP
(855-13-15), 1 P, 6 L (JA 855) [UCLA]. Kingston, 1 Apr 1906, M. Grabham, 1 6, 2 9 [USNM].
Mavis Bank, M. Grabham, 1 6 [USNM]. Mt. Salus Road, elev. 1500 ft, 11 Sept 1967, W.A. Page,
1 pd (777-100), 2 IP (777-10,11), 1 Ipd (779-10), 1 pd (779-104), 2 p? (779-101,105), 1 2, 1 P,
2 L (JA 779); 24 July 1967, W.A. Page (JA 848), 1 IP (848-10), 1 P, 1 L, 11 [UCLA]. Newcastle,

Berlin: Subgenus Howardina 37

elev. 2500-4000 ft, 5 Sept 1964, W.A. Page (JA 124), 1 Ipd (124-15), 3 lp? (124-10,12,14), 1 IP
(124-11), 2 L; 29 Oct 1966, O.G.W. Berlin (JA 648), 1 lpd (648-16), 1 lp? (648-19), 2 pd (648-
100,101), 9 IP (648-10-15,17,18,20), 1 P, 9 L, 13 1; track to Catherines Peak, elev. 4000-4300 ft,
8 Nov 1966, O.G.W. Berlin, 1 P, 1 p, 1 L, 11 (JA 678), 9 2 (JA 679), 2 p? (680-100,101), 1 9,2 P,
1 p, 2 L(JA 680), 1 p,2L,51(JA 681), 1 P, 16 L(JA 682), 1 P, 1 L, 11(JA 683) [UCLA]. Stony
Hill, elev. 1100 ft, 18 July 1964, L. Paton (JA 85), 3 pd (85-101,103,104), 8 p? (85-100,102,105-
108,110,111), 1 P, 3 L;1 Sept 1965, W.A. Page (JA 318), 2 Ipd (318-32,35), 1 Ip? (318-38), 5 pd
(318-33,34,110,111,113), 6 p? (318-102,104,106-109), 1 IP (318-36), 1 6,8 P, 1 p,5 L [UCLA];
1934, 3 2 [USNM]. Uester Road, elev. 4500 ft, 8 Sept 1965, J.N. Belkin, 1 9 (JA 336), 5 lpd (337-
10,14,15,18,24), 2 lp? (337-11,16), 2 pd (337-22,23), 1 p? (337-27), 1 IP (337-17), 5 6,3 9, 1 P,
4 L (JA 337) [UCLA]. Waterworks track, elev. 1100 ft, 8 Sept 1965, J.N. Belkin (JA 335), 2 Ip
(335-10,20), 3 lp? (335-11,22,30), 1 pd (335-100), 1 IP (335-25), 8 6, 2 2, 2 p,6 L [UCLA].
Woodcutters Gap, elev. 4500 ft, 3 Sept 1964, W.A. Page (JA 122), 1 lp? (122-11), 2 d(JA 122),
1 Ipd (123-10), 1 Ip? (123-11), 1 pd (123-100) [UCLA]. St. Ann: Moneague, elev. 500-1750 ft, 9
Sept 1965, D.C. Watson, 2 p? (606-100,101), 4 IP (606-10-13), 4 L (JA 606), 1 2? (JA 607); 17 Aug
1967, J.N. Belkin and W.A. Page, 2 L (JA 733); 31 Aug 1967, W.A. Page, 2 L (JA 762); 2.0 miS of
Moneague near Unity, elev. 1750 ft, 28 Aug 1967, W.A. Page (JA 758), 1 Ipd (758-22), 1 Ip? (758-

21), 1 pé (758-102), 1 IP (758- 20): 29.0 mi post on Moneague Road, elev. 1750 ft, 30 Aug 1967,

W.A. Page, 5 Ipd (763-10,11,13,15,17), 1 lp? (763-16), 1 pd (763-91), 4 p? (763-93,104,110 113),
36,19%,3p,4L; TIGA 763). 2 Ip? (764-11,20), 2 L(JA 764); Moneague-Claremont Road, elev.
750-1250 ft, 17 Sept 1965, W.A. Page, 1 6 (374-100), 1 Ipd (379-12), 2 lp? (379-10,14), 1 IP (379-
11), 4 LIA 379), 11(JA 381); 5 Aug 1967, W.A. Page, 1 p (JA 767) [UCLA]. St. James: Monte-
go Bay, elev. 500 ft, 6 Nov 1966, O.G.W. Berlin and D.C. Watson (JA 669), 1 p? (669-103), 1 IP
(669-20), 2 L [UCLA]; Jan 1940, W.H.W. Komp, 21 [USNM]. Reading, elev. 200 ft, 6 Nov 1966,
O.G.W. Berlin, 1 1(JA 670), 11(JA 671), 2 9 (JA 672) [UCLA]. St. Mary: Broadgate, elev. 500 ft,
25 Aug 1965, D.C. Watson, 1 L (JA 257); 27 Oct 1966, O.G.W. Berlin and D.C. Watson (JA 643),
2 Ipd (643-10,14), 4 Ip? (643-11-13,15), 1 pd (643-109), 2 p? (643-110,111), 1 IP (643-16), 2 p,
1 L, 81 [UCLA]. Castleton, elev. 600 ft, 27 Oct 1966, O.G.W. Berlin and D.C. Watson (JA 644),
1 Ip? (644-20), 1 p? (644-110), 6 P,4 L [UCLA]. St. Thomas: Barrett’s Gap, 14 Nov 1946, G.A.

Thompson, 1 ? [IJ]. Norris, elev. 500 ft, 30 Oct 1966, O.G.W. Berlin and D.C. Watson (JA 653),

1 Ip? (653-21), 1 p? (653- 100), 2 IP (653 -20,22), 1 P, 4 p, 1 L, 1 1 [UCLA]. Pleasant Hill, elev.
250 ft, 14 June 1931, Kisliuk, 7 2 [USNM]. Port Morant Bay, 16 Aug 1967, J.N. Belkin and W.A.

Page, 2 6,1L(JA 729) [UCLA]. Scotland Gate, elev. 500 ft, 30 Oct 1966, O.GW. Berlin and D.C.
Watson (JA 654), 5 pd (654-100,102,103,107,109), 5 p? (101,104-106,108), 10 IP (654-10-19),

1 ¢,10P, 101 [UCLA]. Windsor Forest, elev. 750 ft, 30 Oct 1966, O.G.W. Berlin and D.C. Watson,

1 Ipd (650- 10), 1 pd (650-113), 1 p? (650- oe 4 IP (650-12- 15), 1 P,1L,21(QA 650), 4 Ipd
(651-10,11,14,15), 5 lp? (651-12,17,19-21), 7 pd (651-100-102,105- 108), 5 pe (65 1-103,104,109,
115 116), 3 IP (651-13,16,18), 4 P, 1 p,6 L,11€JA 651) [UCLA] . Westmoreland: Whitehouse,
elev. 100 ft, 11 Sept 1965, W.A. Page, L& (JA 359), 3 2 (JA 360) [UCLA] . Parish not specified:

Sandy Folly, Feb 1945,G A. Thompson, 1 L; Jan 1940, W.H.W. Komp, 1 6; 1940, W.H.W. Komp,
1 9; Jan 1946, G.A. Thompson, a a 1940, 1 ?;M. Grabham, 1 6, 3 2, 1p, 51; R.B. Hill, 3 2;

G.A. Thompson, 1 2 [USNM]; W.A. Page, 1 IP (771-11), 1 Ipé (772. 20), 1 lp? (772-21), 3 L. a
(JA 772), 2 lpé (787-12,14), 2 Ip? (787-11,13), 3 pd (787-108,109,111), 7 p? (787-100-102, 104,

105,107,112), 1 IP (787-10), 1 P, 1 p, 10 L(JA 787) [UCLA].

7. Aedes (Howardina) argyrites Dyar and Nunez Tovar
Figs.3,8
1927.Aedes (Howardina) argyrites Dyar and Nunez Tovar, 1927:190. TYPE: Holotype ?, Vene-
zuela, M. Nunez Tovar [USNM].

Aedes (Howardina) argyrites of Dyar (1928:230); Edwards (1932:155); Lane (1939:99); Anduze
(1941:14); Lane (1953:708); Horsfall (1955:471); Stone and Knight (1956:226); Stone, Knight

38 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

and Starcke (1959:173); Belkin, Schick and Heinemann (1965:74); Cova Garcia, Sutil and
Rausseo (1966a:58; 1966b:141,333).

FEMALE. Apparently distinct from walkeri in the following features (fig.8). Sides
and venter of head mottled with black and white scales; inner dorsocentral mesonotal
line faint, narrow, dotted; midtarsal segment 1 with complete basal ring. The single
known specimen has the apex of the proboscis, palpi, antennae and segments 3-5 of
hindtarsi broken; hindtarsal segments 1 and 2 have basal white rings.

MALE, PUPA, LARVA unknown.

SYSTEMATICS. The pattern of ornamentation of the single known adult of argy-
rites is so similar to that of walkeri that I have no hesitation in assigning argyrites to
the Walkeri Group. The 2 forms are apparently differentiated primarily by the diag-
nostic characters given above but since the holotype of argyrites is damaged it is pos-
sible that additional differences may be present, especially in tarsal markings. A slide
of male genitalia [USNM] from an unspecified locality in Venezuela (P.J. Anduze)
probably represents another specimen of argyrites, as identified by W.H.W. Komp.
There appear to be some differences between these and the male genitalia of walkeri
but without additional material and the absence of the whole adult I hesitate to de-
scribe them at this time. ,

The exact type locality of argyrites is not known but the most likely place is
somewhere in the Cordillera de la Costa in the state of Aragua, the origin of the bulk
of the material collected by Nunez Tovar. Failure to find additional material of argy-
rites in this relatively well-known area suggests that argyrites may be a rare relict
form.

BIONOMICS. The holotype was presumably obtained in a biting-landing collec-
tion. Immature stages of argyrites should be looked for in bromeliad axils, treeholes
and broken bamboo, the known breeding sites of the closely related walkeri.

Aurites Group

FEMALES. Head: Narrow decumbent scales yellowish. Palpus with a few white
scales on apex. Thorax: All light mesonotal lines golden, varied in development; lat-
eral line sometimes moderately to rarely strongly broadened but not silvery. Acros-
tichal bristles not developed on disc. Scutellum with linear golden scales on lateral
lobes. Ssp with or without scales; upper stp patch separate or united. Legs: Foretar-
sus completely dark; midtarsus with basal white markings on segments 1 or 1,2; hind-
tarsus with basal white markings on segments 1,2 or 1-3. Abdomen: Basomedian ter-
gal light markings when present restricted to distal segments.

MALES. Palpus entirely dark. Larger claw of foreleg and midleg with small acute
basal tooth and large submedian tooth; smaller claw with basal tooth only.

MALE GENITALIA. Claspette with a distinct outstanding apical fingerlike cylin-
drical process more than twice as long as wide and bearing a short apical bristle.
Clasper about 0.5 of sidepiece; basal half with minute spicules; subapical hairs absent.
Aedeagus subquadrate, slightly wider at middle, subequal to proctiger in length.

PUPAE. Cephalothorax: Hair 5-C at least 4-branched, moderately long but usual-
ly not reaching base of trumpet; hair 7-C with 3 or more branches; hair 10-C usually
at least 10-branched. Abdomen: Hair 1-II-[V with many branches, 1-II with about as
many branches as float hair (1-I); hair 2-II always mesad of hair 3; hair 5-IV,V usual-
ly double and about 1.5 of tergite following; hair 9-II-VI usually at level or slightly

Berlin: Subgenus Howardina 39

cephalad of hair 6; 9-VII subequal to hair 6-VII; hair 4-VIII double. Paddle: Moder-
ate; practically unpigmented; apex rounded; marginal spicules indistinct.

LARVAE. Head: Hair 1-C unbranched; 4-C short; 5-C single or branched, long;
6-C at least 6-branched, moderate; none of ventral hairs stellate; 14-C single; 15-C
short, not extending cephalad of base of mental plate. Antenna: Shaft with a few
scattered spicules; hair 1-A spikelike, single. Thorax: Hair 13-P absent. Only hairs 0,
1,3,14-P and 5-T stellate, with a few, weakly barbed branches; hair 1-T far mesad of
2,3-T. Abdomen: Integument glabrous; all hairs except 0,8,14 at about same trans-
verse level. Only hairs 2,5-I-VI, 11-I, 7-9-II-VI stellate, with a few weakly barbed
branches; hairs 2 and 9 stronger than others; 9-I small, single; hair 8-VII distinctly
cephalodorsad of 9-VII. Segment VIII: Comb scales in a single row; individual scales
long and fringed to apex. Siphon: Index 5.0-8.5; integument glabrous; acus absent
but base of siphon with small lateral indentation. Pecten not reaching middle of si-
phon; teeth long, without denticles. Anal Segment: Saddle body without distinct
spicules except distally; caudal margin with row of long spines; hair 1-X always sin-
gle and at lower end of caudal margin. Ventral brush with 6 pairs of hairs on a weak
boss without grid bars.

DISCUSSION. The clearly marked Aurites Group shows a phenomenon unique in
the subgenus: a flock of 4 sibling species sympatric in one small area. To date the
group has been recognized only in Jamaica, Were some of the species are quite com-
mon (fig.3).

There is a suggestion of some affinity of this group with the sympatric Walkeri
Group in the general similarity in the mesonotal pattern of the adults. However, the
immature stages of the 2 groups are very distinct. There is a much greater resem-
blance in the mesonotal pattern of the adults with the Sexlineatus Section except for
the prescutellar line, and the larvae and pupae also resemble the Sexlineatus Section
more than any group in the Walkeri Section. Although this resemblance may be en-
tirely fortuitous, 2 more plausible explanations are possible. First, the Aurites Group
might be the primitive stock of the Walkeri Section derived from the Sexlineatus Sec-
tion; this is very unlikely since the entire Sexlineatus Section lacks several important
primitive characters represented in other groups of the Walkeri Section (acrostichal
bristles of the adults and comb scales in a patch in the larva). Second, the view held
here; the Aurites Group may have been formed through relatively recent introgres-
sion between the Sexlineatus Section and the Walkeri Group. This interpretation
would also account for the presence of a species flock in the Aurites Group, since
such radiation usually takes place soon after a new phyletic line is established. It is
also supported indirectly by the presence in Jamaica of several species in other gen-
era that are either identical with or very closely related to Central American species,
strongly suggesting a relatively recent connection between these areas.

No specific differences in the male genitalia of members of the group have been
found and the adults of inaequalis and stenei cannot be separated from each other.
In other respects the 4 species are readily differentiated from one another as indi-
cated in the keys.

There appear to be some ecological barriers separating 2 of the species. A. inae-
qualis breeds in treeholes, broken bamboo and rockholes, largely at low elevations,
and never in bromeliad axils, which are utilized exclusively by the other 3 species.
A. aurites breeds primarily at elevations above 3,500 feet and has never been found
associated with the other bromeliad-breeding members of the group. A. grabhami and
stenei occur primarily below 2,500 feet but grabhami has been collected at elevations
up to 4,000 feet. These 2 species are usually found associated in collections but it is
not known if they occupy the same axils under natural conditions.

40 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

8. Aedes (Howardina) aurites (Theobald)
Figs.3 ,8,19,20

1907.Howardina aurites Theobald, 1907:216-218. TYPE: Lectotype ?, Newcastle (St. Andrew),
Jamaica, elev. 4000 ft, July 1906, Col. Loscombe [BM; selection of Belkin, 1968].

Aedes (Howardina) aurites in part of Dyar (1928:236); Edwards (1932:155); Lane (1939:99; 1953:
720-722); Horsfall (1955:471).

Aedes aurites of Howard, Dyar and Knab (1917:859-860); Dyar (1918:73); Johnson (1919:424);
Gowdey (1926:73).

Aedes aurites in part of Hill and Hill (1948:29,42).

Howardina aurites of Theobald (1910:220); Surcouf and Gonzalez-Rincones (1911:139-140).

Aedes (Howardina) inaequalis in part of Thompson (1956:199-200); Stone, Knight and Starcke
(1959:174); Belkin, Schick and Heinemann (1965:31); Forattini (1965 :399,400).

Aedes inaequalis in part of Stone and Knight (1956:219).

FEMALE. Wing: 2.6 mm. Proboscis: 2.00 mm. Forefemur: 1.8 mm. Abdomen:
about 2.7 mm. Head: Light decumbent scales golden on median stripe and short or-
bital line, yellowish on sides and venter. Erect scales yellowish, a few lateral ones
darker. Palpus about 0.16 of proboscis. Thorax (fig.8): Mesonotal dark scales dense;
light lines golden, moderately broad, wider than dark lines; inner dorsocentral line
reaching to prescutellar space, followed by moderately broad median prescutellar
line; outer dorsocentral line complete, merging with lateral marginal line anteriorly;
lateral marginal line long, extending from side of anterior promontory to wingbase,
very broad from scutal angle to wingbase due to presence of golden scales on antealar
area above paratergite. Median scutellar lobe with a row of narrow golden scales
flanked on sides by broad dark ones; lateral lobes with a few narrow, golden and
black scales. Apn with a distinct silvery patch; ssp bare; patches on base of pra, up-
per and lower stp merged and continuous; rest of patches as in the section and group.
Legs: Forefemur and midfemur predominantly dark anteriorly except for a few basal
creamy scales, posterior surface and hindfemur predominantly creamy except for dis-
tal dark areas. Midtarsus with few white scales on segment 1; hindtarsus with short
white rings on 1,2, rest dark scaled. Haltere: Stem pale, knob predominantly dark,
fringe scales whitish. Abdomen: Basomedian tergal light scaling inconspicuous. Ster-
nites II-IV straw yellow, V predominantly yellow with few apical dark scales; rest of
segments with basal creamy and distal dark scales.

MALE. Wing: 2.34 mm. Proboscis: 2.16 mm. Forefemur: 1.71 mm. Essentially
similar to female in coloration.

MALE GENITALIA (fig.19). As figured; diagnostic characters as in the key. Ninth
tergite lobe indistinct, with 2 strong bristles. Sidepiece length about 3.0 median
width. Clasper about 0.6 of sidepiece, basal 0.3 with minute spicules; sly ae short,
about 0.25 of clasper.

PUPA (fig.19). Abdomen: 2.48 mm. Trumpet: 0.38 mm; index 6.0-6.2. Paddle:
0.7 mm. Readily separated from inaequalis and stenei by the absence of lightly pig-
mented area on metanotum and hair 9-C single; from grabhami by hair 2-VI laterad
of 3-VI. Cephalothorax: Mesonotum and metanotum uniformly lightly pigmented;
trumpets contrasting. Hair 1(1), 2(2, 1-2), 3(1, 1-2), 4(5, 3-8), 5(5, 4-7), 6(1, 1-2),
7(4, 4-6), 8(1), 9(1), 10(10, 9-16), 11(1), 12(3, 2-6). Abdomen: Tergite I uniformly
lightly pigmented except for usual membranous areas; rest of segments light, pigmen-
tation progressively lighter caudad. Segment I: hair 1(12-18 primary branches, each
with many secondary ones), 2(1), 3(1), 4(6, 3-8), 5(6, 2-6), 6(1, 1-2), 7(2, 2-3), 9(1),

Berlin: Subgenus Howardina 41

10(1, present in some). Segment II: hair 0(1), 1(37-56 branches, variable in different
specimens), 2(1), 3(1), 4(4, 2-4), 5(1), 6(1), 7(4, 3-4), 9(1), 10(1, present in some).
Segment III: hair 0(1), 1(14, 6-24), 2(1), 3(1, 1-3), 4(4, 2-4), 5(1), 6(1), 7(3, 2-4),
8(3, 2-4), 9(1, 1-2), 10(2, 2-3), 11(1), 14(1). Segment IV: hair 0(1), 1(8, 5-16), 21,
1-2), 3(4, 3-6), 4(2, 1-4), 5(1, 1-3), 6(1), 7(3, 2-5), 8(2, 1-3), 9C1), 10@, 2-4), 11Q1), _
14(1). Segment V: hair 0(1), 1(2, 1-7), 2(1, 1-2), 3(2, 2-3), 4(5, 3-8), 5(1), 6(1), 744,
2-6), 8(2, 1-3), 9(1), 10(€1, 1-2), 11(1), 14(1). Segment VI: hair 0(1), 1(1, 1-6), 2(1),
3(2, 1-3), 4(4, 2-6), 5(2, 1-5), 6(1, 1-3), 7(1), 8(3, 1-8), 91), 10(1), 1101), 14().
- Segment VII: hair 0(1), 1(1), 2(1), 31, 1-2), 4(2, 1-2), 5C1, 1-2), 6(2, 2-4), 7(1, 1-2),
8(3, 2-4), 9(3, 2-4), 10(1), 11(1), 14(1). Segment VIII: hair 0(1), 4(2), 9(10, 8-12),
14(1). Paddle: Almost without pigmentation; midrib faintly pigmented; external
margin smooth. Male genital lobe extending to 0.42 of paddle, female genital lobe to
O27:

LARVA (fig.20). Head: 0.73 mm. Siphon: 0.97 mm. Anal Saddle: 0.24 mm. Read-
ily separated from stenei by hair 2-VII small and single; from inaequalis by the longer
subequal gills and at least 3-branched hair 5-C; from grabhami by the 3-branched
hair 5-C. Chaetotaxy as given below, hairs moderately pigmented. Head: Width about
1.18 of length; pigmentation light, deeper around mouthparts and behind ocular area
ventrally. Mental plate with a strong median tooth and 10 distinct teeth on either
side. Hair 0(1), 1(1), 3(1), 4(4, 4-5), 5(3, 2-4), 6(10, 9-13), 7(6, 4-7), 8(1), 9(3, 2-3),
10(1), 11(3, 3-5), 12(1, 1-2), 13(1), 141, small), 15(3, 3-4). Antenna: Length about
0.44 of head, very lightly pigmented and with few scattered spicules. All hairs single;
hair 1-A about 0.5 from base. Thorax: Long hairs attenuate apically and weakly
barbed; stellate hairs with a few weakly barbed branches. Prothorax: hair O(11, 8-
14), 109, 9-11), 21), 313, 12-14), 41), 5Q, 2-3), 6(1), 7(1, 1-2), 8(8, 8-12), 9Q, 1-
3), 10(1), 11(4, 2-5), 12(1), 14(16, 11-19). Mesothorax: hair 1(5, 3-6), 2(1), 3(1),
4(1), 5(1), 6(5, 4-6), 7(1), 8(7, 6-8), 9(8, 6-9), 10(1), 11(2, 1-3), 121), 13(10, 6-16),
14(5, 5-9). Metathorax: hair 1(2, 1-4), 2(1), 3(4, 3-5), 4(7, 3-9), 5(9, 6-10), 6(1),
7(7, 6-9), 8(4, 3-5), 9(5, 4-7), 10(1), 11@2, 1-2), 121), 13(7, 4-7). Abdomen: Stellate
hairs as on thorax. Segment I: hair 1(3, 2-3), 2(7, 6-8), 3(1), 4(4, 2-5), 5(7, 7-10),
6(3, 3-4), 7(1), 9C1, 1-2), 10(1), 11(7, 6-9), 13(5, 3-6). Segment II: hair 0(1), 1(@2, 1-
3), 2(7, 6-8), 3(1), 4(2, 2-4), 5(9, 7-10), 6(3), 7(3, 2-4), 8(1, 1-2), 9(6, 3-7), 10(1),
111), 12(1), 133, 3-5). Segment III: hair 0(1), 1(2, 2-3), 2(6, 5-8), 3(1), 4(@2, 2-3),
5(10, 8-10), 6(1, 1-2), 7(5, 4-7), 8(1), 9(6, 5-6), 10(1), 11(1), 12(1), 13(5, 3-6), 14(1).
Segment IV: hair 0(1), 1(2, 2-3), 2(7, 6-7), 3(1), 4(2, 2-3), 5(10, 8-11), 6(1, 1-2), 7¢4,
4-8), 8(1, 1-2), 9(6, 5-7), 10(1), 11€1), 121), 13(5, 4-6), 14(1). Segment V: hair
O(1), 1(2), 2(6, 5-7), 3(1), 4(2, 2-5), 5(9, 7-10), 6(1), 7(3, 3-5), 8C1, 1-2), 9(6, 5-6),
101), 11(€1), 121), 13(5, 4-7), 14(1). Segment VI: hair 0(1), 1(2), 265, 4-6), 3(1),
4(1), 5(6, 5-9), 6(1), 7(2, 2-3), 8(3, 1-3), 9(4, 4-6), 10(1), 11(1), 121), 13(7, 5-8),
14(1). Segment VII: hair 0(1), 1€1, 1-2), 2(1), 3(1), 41), 5(3, 2-4), 6(3, 2-3), 7(1),
8(6, 4-7), 9(3, 2-3), 10(1), 11(1), 12(1), 133, 2-5), 14(1). Segment VIIT: Comb scales
in a single row, 18(16-19) in number; individual scale long and fringed uniformly to
apex. Hair 0(1), 1(4, 3-5), 2(1), 3(6, 4-6), 4(1), 5(2, 2-3), 14(1). Siphon: Index about
6.0-6.5; moderately pigmented, with darker basal ring. Pecten extending to about
0.45, with 19(14-21) teeth; proximal teeth smooth, distal with a few apical spinules.
Hair 1(2, 2-3), 2(1), 3-5 not visible, 6(1), 7(1), 8(2, 1-3), 9(1). Anal Segment: Saddle
moderately pigmented; caudal margin with spines of varying sizes. Gills subequal,
long, about 2.0 of saddle length. Hair 1(1), 2(5, 4-7), 3(1), 4a(4, 4-5), 4b(4), 4c(4,
3-5), 4d(4, 3-5), 4e(4, 3-4), 4f(3, 2-4).

SYSTEMATICS. A. aurites is undoubtedly a distinct species erroneously synony-

42 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

mized by Thompson (1956:199) with inaequalis. It is readily differentiated in the
adult stage from all the other members of the group, as pointed out by Theobald
(1907:217) in the original description, by the large continuous silvery patch on ssp.
There are other diagnostic features in the adults as indicated above and the immature
stages are readily separated from the other members of the group.

There is no striking variation in adult ornamentation in the small series of speci-
mens examined and the diagnostic characters of the immature stages are quite con-
stant.

BIONOMICS. The immature stages of aurites have been found only in arboreal
bromeliads, in my experience generally at elevations of more than 3,500 feet and
never in association with grabhami or stenei, the other bromeliad-breeding members
of the Aurites Group. It appears that this species occurs at lower elevations at least
occasionally as Thompson reared several adults from larvae collected at Hope Farm
[USNM] at an elevation of 1,000 feet in 1946, and 2 females (JA 702) were caught
recently (1967) near Kingston at elevations of about 1,000-1,500 feet. Females have
been collected biting humans during the day but no attempt has been made to study
their activity cycle. From the available data, aurites appears to have a restricted geo-
graphic range, having been collected only from the parishes of Manchester and St.
Andrew.

DISTRIBUTION (fig.3). Endemic to the island of Jamaica where it has been re-
ported from the parishes of Manchester, St. Andrew and Kingston, primarily at ele-
vations of 3400 to 5250 ft. Material examined: 145 specimens; 24 6, 33 9, 44 pupae,
44 larvae; 36 individual rearings (22 pupal, 7 larval, 7 incomplete).

JAMAICA. Manchester: Hope Farm, 1 Jan 1946, G.A. Thompson, 2 Ipd (460101-24,26), 1 1d
(460101-30), 1 lp? (460101-27), 1 lp (460101-15); 1 Oct 1946, G.A. Thompson, 1 p (461001-1)
[USNM]. St. Andrew and Kingston: Cinchona Gardens, elev. 4000-5250 ft, 8 Sept 1965, J.N. Bel-
kin, 1 d (JA 341); 24 Aug 1967, W.A. Page, 3 L (JA 752) [UCLA]. Ferry River, 4 Aug 1967, W.A.
Page, 2 (JA 702) [UCLA]. Hardwar Gap, elev. 3800 ft, 1 lpd (647-20), 2 pd (647-110,111)
[UCLA]. Kingston, July 1905, M. Grabham, 1 ? [USNM]. Newcastle, elev. 4000 ft, 5 Sept 1964,
W.A. Page (JA 125), 1 lp? (125-10), 2 pd (125-102,106), 7 p? (125-100,101,103-105,107,114),
1 IP (125-11), 1 6, 7 2, 4 L; track to Catherines Peak, elev. 4000-4300 ft, 8 Nov 1966, O.G.W. Ber-
lin, 2 pd (677-101,102), 1 p? (677-100), 5 2 (JA 679), 2 pd (680-110,111), 1 p? (680-112), 3 6,
1 P (JA 680), 1 Ipod (681-10), 1 pd (681-100), 1 6, 2 9, 2 p, 7 L(JA 681), 1 pd (682-101), 3 p?
(682-100,102,103), 3 IP (682-10-12), 3 6,5 P, 13 L, 21(JA 682). Uester Road, elev. 3400-4500 ft,
8 Sept 1965, J.N. Belkin, 2 IP (337-12,13), 1 6, 1 2 (JA 337), 1 L(JA 338) [UCLA].

9. Aedes (Howardina) grabhami Berlin, n.sp.
Figs.3,7,8,21,22

TYPES: Holotype ¢ (JA 688-20) with associated pupal and larval skins and genitalia slide, near
Hope Farm, Porus, Manchester, Jamaica, elev. 1000 ft, 12 Nov 1966, O.G.W. Berlin [USNM]. Al-
lotype ? (JA 688-11) with associated pupal and larval skins, same data as holotype [USNM]. Para-
types: 11 6 (JA 688-13,100,101,107,108,111,114-116,126,128), 14 9 (JA 688-12,15,102,103,
110,113,120-123,129-132), 1 IP (JA 688-16), 1 6,5 p, 8 L, 1 1, same data as holotype [USNM,
BM, UCLA].

1906.Howardina aureostriata Grabham, 1906:171-173. TYPES: 6, °, larva, Newcastle (St. An-
drew), Jamaica, elev. 4000 ft, Sept 1905, Loscombe and Maclaverty [NE].

Aedes (Howardina) aureostriata in part of Bonne and Bonne-Wepster (1925 :370-371).
Aedes aureostriata of Gowdey (1926:73).

Berlin: Subgenus Howardina 43

Aedes aureostriata in part of Howard, Dyar and Knab (1917:855-859); Dyar (1918:73,80).

Howardina aureostriata of Theobald (1910:609-611).

Haemagogus aureostriata of Dyar and Knab (1906a:166,167).

Aedes (Howardina) inaequalis in part of Thompson (1947:78; 1956:199-200); Stone, Knight and
Starcke (1959:174); Belkin, Schick and Heinemann (1965:31); Forattini (1965 :399,400).

Aedes inaequalis in part of Stone and Knight (1956:219).

Aedes (Howardina) aurites in part of Dyar (1928:236); Edwards (1932:155); Lane (1939:99;
1953:720-722); Horsfall (1955:471).

Aedes aurites in part of Hill and Hill (1948:29,42).

FEMALE. Wing: 2.15 mm. Proboscis: 1.68 mm. Forefemur: 1.33 mm. Abdomen:
about 2.1 mm. Essentially similar to aurites, differing mainly in the following. Head:
Erect scales entirely dark. Thorax (fig.8): Mesonotal dark scales not as dense as in
aurites; light lines usually distinctly narrower than in aurites, in a few specimens
broader than in aurites and rarely almost completely fused; outer dorsocentral line
usually distinctly shorter, not merging with lateral marginal line anteriorly, in a few
specimens merging with lateral marginal line; lateral marginal line long, moderately
broad, extending from side of anterior promontory to wingbase. Antealar area above
paratergite predominantly with scattered golden and a few bronzy scales. Lateral scu-
tellar lobes with a few narrow golden scales only. Apn with a moderate silvery patch;
ssp with a distinct silvery patch. Patches on base of pra, upper and lower stp distinct,
separate. Legs: Midtarsus with few white scales on segments 1,2; hindtarsus with
moderate white rings on 1-3. Haltere: Stem pale, knob predominantly white but
with a few dark scales dorsally. Abdomen: Sternites II, III straw yellow, [V predomi-
nantly straw yellow, a few apical scales dark; rest of segments with basal yellow and
distal dark scales.

MALE. Wing: 2.21 mm. Proboscis: 1.75 mm. Forefemur: 1.31 mm. Essentially
similar to female in color.

MALE GENITALIA (fig.21). As figured; diagnostic characters as in the key. Ninth
tergite lobe indistinct, with 2 strong bristles. Sidepiece length about 3.0 median
width. Clasper length 0.65 of sidepiece, basal 0.33 with minute spicules; spiniform
about 0.2 of clasper.

PUPA (fig.21). Abdomen: 2.75 mm. Trumpet: 0.38 mm; index 4.6-4.8. Paddle:
0.69 mm. Chaetotaxy as given below, hairs moderately pigmented. Readily separated
from inaequalis and stenei by the absence of lightly pigmented area on metanotum
and abdominal segments uniformly lightly pigmented; from aurites by the hair 2-VI
distinctly mesad or at level of 3-VI. Cephalothorax: Mesonotum and metanotum
lightly pigmented. Hair 1(1), 2(2, 2-3), 3(1), 4(7, 4-8), 5(5, 4-7), 6(1, 1-2), 7(4, 3-5),
8(1, 1-3), 9(1, 1-2), 10(22, 18-32), 11(1), 12(5, 3-6). Abdomen: Lightly pigmented,
progressively lighter caudad. Segment I: hair 1(12-16 primary branches, subequal in
length with many secondary branches), 2(2, 1-3), 3(1), 4(8, 6-8), 5(8, 4-8), 6(1, 1-2),
7(4, 3-4), 9(1), 10€1), 11€1). Segment II: hair 0(1), 1(60, 50-74), 2(1), 3(1), 465, 4-
6), 5(1), 6(1), 7(5, 4-6), 9(1). Segment III: hair 0(1), 1(13, 8-17), 21), 3d, 1-3), 4¢4,
3-6), 5(1), 6(1), 7(4, 3-6), 8(4, 2-4), 9(1), 10(2, 2-3), 11(1), 141). Segment IV: hair
O(1), 109, 8-14), 2(1), 3(4, 4-6), 4(4, 2-4), 5(1), 6(1), 7(4, 3-6), 8@2, 1-4), 9(1), 10@,
1-2), 11(1), 14(1). Segment V: hair 0(1), 1(1), 2(1), 3@, 1-3), 4(4, 4-7), 5(1), 6(1),
7(5, 4-6), 8(2, 2-3), 9(1), 10(1, 1-2), 11(1), 141). Segment VI: hair 0(1), 1(1, 1-2),
2(1, 1-2), 3(2, 1-2), 4(4, 3-4), 5(1), 6(1), 7(1), 8@, 2-3), 9€1), 10€1, 1-2), 11(1),
—14(1). Segment VII: hair 0(1), 1€1), 2(1), 3(2, 1-2), 4(2), 5(1), 6(2, 2-5), 7(1), 8G,
2-4), 9(2, 2-4), 10(1), 11€1), 14(€1). Segment VIII: hair 0(1), 4(2, 1-2), 9(10, 8-11),

4d Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

14(1). Paddle: Almost without pigmentation, midrib faintly pigmented. Male genital
lobe extending to 0.42 of paddle, female genital lobe to 0.35.

LARVA (fig.22). Head: 0.7 mm. Siphon: 0.92 mm. Anal Saddle: 0.22 mm. -Chae-
totaxy as given below, hairs lightly to moderately pigmented. Readily separated from
stenei by the small and single hair 2-VII; from aurites by hair 5-C always single; from
inaequalis by hair 5-C single and longer, narrow gills. Head: Width about 1.14 of
length. Mental plate with a strong median tooth and 10 distinct teeth (9-12) on each
side, often: a small detached tooth at base. Hair 0(1), 1(1), 3(1), 4(3, 2-5), 5(1), 6(6,
5-10), 7(3, 2-3), 8(1), 9@, 2-3), 10(2, 1-2), 11(2, 1-7), 12(@2, 2-4), 131), 14(1), 15@G,
2-4). Antenna: Length about 0.49 of head, very lightly pigmented and with a few
scattered spicules. Hair 1-A at about 0.46 from base. Thorax: Stellate hairs with a
few, weakly barbed branches; long hairs barbed, with attenuate apices. Prothorax:
hair 0(7, 4-12), 1(6, 5-9), 2(1), 3(13, 8-19), 4(1, 1-2), 5(2, 1-2), 61), 71), 8@, 2-11),
9(2, 1-2), 10(1), 11@2, 2-4), 121), 14(13, 9-22). Mesothorax: hair 1(2, 2-13), 2(1),
3(1), 4(1), 5(1), 6(4, 2-6), 7€1), 8(5, 4-7), 9(7, 6-8), 1001), 11@, 1-2), 121), 1344,
2-4), 14(5, 4-7). Metathorax: hair 1(2, 1-12), 2(1), 3(4, 2-5), 4(2, 2-10), 5(5, 5-14),
6(1), 7(7, 6-8), 8(5, 4-8), 9(6, 6-9), 10(1), 11(2), 121), 13(4, 3-10). Abdomen: Long
and stellate hairs as on thorax. Segment I: hair 1(3, 2-14), 2(6, 6-14), 3(1), 4(4, 3-4),
5(4, 4-12), 6(3), 7(1), 9(1), 10(1), 1105, 5-11), 13(4, 4-11). Segment II: hair 0(1),
1(2, 2-12), 2(5, 5-13), 31), 4(2, 2-4), 5(7, 5-14), 6(3, 2-3), 7(2, 2-3), 8(1), 9(5, 5-9),
101), 11€1), 12(1), 13(3, 2-8). Segment III: hair 0(1), 1(2, 2-3), 2(5, 5-11), 31),
4(2, 2-3), 5(7, 5-12), 6(1, 1-2), 7(6, 3-7), 8(1), 9(5, 4-9), 10€1), 11€1), 121), 13@,
2-8), 14(1). Segment IV: hair 0(1), 1(2, 1-3), 2(5, 5-12), 3(1), 4Q, 2-3), 5(6, 6-14),
6(1), 7(6, 5-7), 8(1), 9(5, 5-9), 10(1), 11€1), 121), 13(3, 3-8), 14(1). Segment V:
hair O(1), 1(2, 2-3), 2(5, 5-10), 3(1), 4(4, 3-5), 5(5, 4-13), 6(1), 7(6, 3-7), 8(1), 965,
5-7), 101), 11€1), 121), 13(3, 2-8), 14(1). Segment VE: hair 0(1), 1@2, 1-3), 2(4, 4-
8), 3(1), 4(1), 5(2, 2-9), 6(1), 7(3, 2-5), 8(3, 2-3), 9(3, 2-7), 10(1), 1101), 12(1), 13,
2-11), 14(1). Segment VII: hair 0(1), 1(2, 1-2), 2(1), 31), 41), 5(3, 2-3), 6(3, 2-5),
71), 8(5, 5-10), 9(€1, 1-2), 10(1), 1101), 121), 131, 1-3), 141). Segment VIIT:
Comb scales 16(15-19), moderately pigmented, elongate, fringed on sides and apex.
Hair 0(1), 1(3, 3-4), 2(1), 365, 4-7), 4(1), 5(2, 2-5), 141). Siphon: Index about 6.8-
8.5; moderately pigmented, with darker basal ring. Pecten extending to about 0.44;
teeth 17(14-22). Hair 1(2, 2-3), 2(1), 3-5 not visible, 6(1), 7(1), 8(2, 1-2), 9(1). Anal
Segment: Saddle moderately pigmented. Gills narrow, length about 1.5 of saddle,
ventral 0.8 of dorsal. Hair 1(1), 2(5, 5-8), 3(1), 4a(5, 4-7), 4b(5, 4-5), 4c(4, 3-5),
4d(4, 4-5), 4e(4, 2-6), 4f(3, 2-4).

SYSTEMATICS. Although it is almost certain that grabhami represents the same
taxon as aureostriatus Grabham, 1907, which is preoccupied in Aedes by aureosStria-
tus (Doleschall, 1857) and aureostriatus (Leicester, 1908), I am proposing it as a new
Species and not as a replacement name for aureostriatus because the types of the lat-
ter are lost and therefore the identity of this nominal taxon cannot be objectively de-
termined. This species was believed to be identical with inaequalis and aurites by
Thompson (1956) who used the former name because of priority. A. grabhami is
readily differentiated from the other species of the Aurites Group in all stages by the
characters indicated in the diagnosis and keys.

There is considerable variation in the ornamentation of adults of grabhami. The
majority have distinct golden mesonotal lines, broader than in inaequalis and stenei
and narrower than in aurites; a few have these lines narrower than in inaequalis or
stenei. The most striking departure is found in a few specimens with the mesonotum
almost completely golden except for a few dark scales in indistinct narrow longitu-

Berlin: Subgenus Howardina 45

dinal lines; in these the hindtarsal light markings are restricted to segments 1 and 2.
I have retained these specimens under grabhami on the basis of similarity in other
diagnostic features but their exact taxonomic status is doubtful. They may represent
a distinct species or possibly hybrids between typical grabhami and stenei which have
been found in the same collections. There is also considerable variation in the chaeto-
taxy of the larvae, some being much more “hairy” than others; however, the diagnos-
tic characters given above apply to all the material I have seen.

BIONOMICS. The immature stages of grabhami are found in both terrestrial and
epiphytic bromeliads. A. stenei is frequently associated with grabhami in the same
bromeliads but it is not known if the 2 species actually occupy the same axils. I did
not find grabhami above 1,800 feet elevation but the original material of aureostria-
tus is reported to have been collected at an altitude of about 4,000 feet. A. grabhami
is, next to stenei, the most common member of the Aurites Group and has been re-
ported from all parishes of Jamaica except Westmoreland, Trelawny and St. Elizabeth.
Females were found to bite humans along with stenei and inaequalis and showed the
same small peak of biting activity at 1700 hrs, as the latter (fig.7).

DISTRIBUTION (fig.3). Endemic to the island of Jamaica where it has been re-
ported from all parishes except Westmoreland, Trelawny and St. Elizabeth at eleva-
tions of 100 ft to 1800 ft. Material examined: 1223 specimens; 172 6, 385 9, 373
pupae, 293 larvae; 298 individual rearings (198 pupal, 69 larval, 31 incomplete).

JAMAICA. Clarendon: Belcarres, 7 Nov 1967, W.A. Page (JA 902), 1 pd (902-101), 1 p? (902-
102), 3 L [UCLA]. Hanover: Lethe, elev. 500 ft, 6 Nov 1966, O.G.W. Berlin and D.C. Watson, 2 L
(JA 667) [UCLA]. Manchester: Hope Farm, 8 July 1945, G.A. Thompson, 1 Ipdé (450708-2), 2 lp?
(450708-9,11), 1 lp (450708-7), 1 1(450708-8); 1 Aug 1945, G.A. Thompson, 1 16 (450801-12),
1 2 (450801-12), 2 1(450801-6,8); 1 Oct 1945, G.A. Thompson, 2 Ip¢ (451001-6,9), 1 16 (451001-
1), 1 Ip (451001-14); Oct 1945, 3 L; 1 Jan 1946, G.A. Thompson, 2 lp? (460101-4,12), 16
(460101-15); 8 Jan 1946, 1 6, 1 9 [USNM]; elev. 1000-1100 ft, 10 Sept 1965, W.A. Page, 1 Ip?
(354-12), 2 pd (354-105,106), 3 p? (354-103,104,107), 1 d (JA 355), 1 9 (JA 356); 12 Nov 1966,
O.G.W. Berlin, 210 2 (JA 684), 2 pé (686-100,102), 1 p? (686-101), 2 Ip? (687-11,12), 5 pd (687-
100-104), 1 IP (687-10), 1 P (JA 687), 2 Ipd (688-13,20), 3 Ip? (688-11,12,15), 10 pd (688-100,
101,107,108,111,114-116,126,128), 12 p? (102,103,110,113,120-123,129,130-132), 1 IP (688-
16), 16,5 p,8L, 11(JA 688, type series) [UCLA]. Mandeville, 5 Mar 1928, Pickering, 5 9 [USNM].
Porus, elev. 900-1000 ft, 23 Oct 1966, O.G.W. Berlin and D.C. Watson, 1 lpé (630-11), 1 lp? (630-
10), 1 L (JA 630), 4 pd (631-105,106,109,112), 2 p? (631-108,113), 29,1 P,2p,1L,11(JA
631), 2 Ipd (632-10,12), 1 Ip? (632-11), 6 pd (632-101,103,104,108,112,114), 5 p? (632-100,102,
105-107), 2 IP (632-13,15), 1 P, 2 p, 19 L, 21 (JA 632); 1 Nov 1966, O.G.W. Berlin, 1 Ipé (655-
10), 6 pd (655-101,103,105-107,109), 4 p? (655-100,104,108,111), 4 IP (655-11-14),4.L,61(JA
655), 2 L (JA 656), 2 pd (658-110,114), 4 p? (658-103,107,112,113), 16,1 p, 1 L, 11(JA 658)
[UCLA]. Portland: Kemney-Sherwood Forest, elev. 750 ft, 12 Aug 1967, J.N. Belkin and W.A.
Page, 1 Ipé (706-11), 1 p? (706-100), 1 p (JA 706), 2 lp? (708-13,16), 1 pd (708-11), 1 Ip (708-
14), 1 @ (JA 708) [UCLA]. St. Andrew and Kingston: Hermitage Dam Road, 1 Sept 1965, W.A.
Page (JA 319), 1 lp? (319-105); 17 Aug 1967, J.N. Belkin and W.A! Page, 2 lpd (723-21 ,23), 1 lp?
(723-24), 1 IP (723-20), 2 p? (725-113A,114), 1 6, 1 P(JA 725); 5 July 1967, W.A. Page and IN.
Belkin (JA 765), 1 pd (765-100); 25 Aug 1967, W.A. Page (JA 854), 1 Ipd (854-13), 2 po (854-
105,106), 1 p? (854-107), 2 IP (854-15,16), 2 L;5 Aug 1967, W.A. Page and J.N. Belkin (JA 892),
4 Ip (892-10,13,15,16), 3 Ip? (892-11,12,14), 2 pd (892-100,104), 3 P, 1 L [UCLA]. Kingston,
M. Grabham, 1 6 [USNM]. St. Ann: Moneague, 17 Aug 1967, J.N. Belkin and W.A. Page (JA 733),
1 pd (733-105), 4 p? (733-40,90,96,114), 2 6, 1 P,6 p, 3 L, 11 [UCLA]; Faiths Pen Schoolyard,
elev. 1750 ft, 31 Aug 1967, W.A. Page (JA 762), 2 pé (762-100,101), 16,19, 1P,11 [UCLA];
2.0 mi S of Moneague near Unity, elev. 1750 ft, 25 Aug 1967, W.A. Page (JA 758), 6 pd (758-100,
103,104,109,110,113), 2 p? (758-30,101), 1 6, 2 9, 3 p [UCLA]; 29.0 mi S of Moneague, elev.
1750 ft, 30 Aug 1967, W.A. Page, 1 6, 2 9, 4 P, 31 (JA 763), 2 lpd (764-10,12), 1 Ip? (764-30),

46 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

9 pd (764-91,94,95,99 103-106,109), 5 p? (764-92,98,100,102,108), 2 d, 12,8 P, 6p, 21LQJA
764) [UCLA] ; Moneague-Claremont Road, 17 Sept 1965, W.A. Page, 1 lpd (375-12), 1 lp? (375-
13), 1 p? (375-100), 7 2 (JA 375), 1 pd (378-100), 1 p? (378-101), 1 IP (378-13), 1 Ipd (380-12),
4 Ip? (380-10,11,13,17A), 3 pd (380-101,104,105), 2 p? (380-100,107), 5 IP (380-14,15,17,18,
20) [UCLA]; Moneague-Ocho Rios Road, 10 Sept 1967, W.A. Page (JA 776), 1 p? (776-101)
[UCLA]. Mt. Diablo, 17 Sept 1965, W.A. Page (JA 376), 1 Ip? (376-12); 17 Aug 1967, J.N. Belkin
and W.A. Page, 6 Ipd (731-21,33,35-38), 1 lp? (731-31), 9 pd (731-90,92,93,105,108-112), 12 p¢
(731-80,81,91,94,97,99,103-107,113), 2 IP (731-30,34), 11 6,79, 2P,17 p, 24 L(VJA 731), 1 Ipd
(732-10), 10 pd (732-70,94,97,100-102,104 111,112,114), 6 p? (732-90,92,93,105 ,107,108), 1 °,
1 P,3 p, 5 L, 21(JA 732), 4 lpd (733-13-16), 1 pd (733-100), 1 IP (733-11), 12 L JA 733), 1 pd
(734-102), 2 p? (734-101,105), 2 P, 1 L (JA 734), 1 2 (JA 735) [UCLA] . St. Elizabeth: Darling-
ton; 28 Oct 1967, W.A. Page, 1 p,4 L(JA 901) [UCLA]. St. James: Five Miles, elev. 100 ft, 6 Nov
1966, O.G.W. Berlin and D.C. Watson (JA 676), 1 Ip? (676-11), 2 pd (676-115,121), 6 p? (676-
104,107,112,113,116,123), 1 IP (676-17), 1 3, 1 2, 3 p, 2 L [UCLA]. Montego Bay, elev. 500 ft,
6 Nov 1966, O.G.W. Berlin and D.C. Watson (JA 669), 1 IP (669-11) [UCLA]. Reading, elev. 200 ft,
6 Nov 1966, O.G.W. Berlin and D.C. Watson, 1 1 (JA 670) [UCLA]. St. Mary: Broadgate, elev.
500 ft, 27 Oct 1966, O.G.W. Berlin and D.C. Watson (JA 643), 2 lpd (643-33,43), 1 lp? (643-37),
1 L,11 [UCLA]. Leith Hall, 13 Nov 1966, O.G.W. Berlin and D.C. Watson, 4 L (JA 689) [UCLA].
Windsor Forest, elev. 750 ft, 30 Oct 1966, O.G.W. Berlin and D.C. Watson, 1 Ip? (650-22), 4 pd
(650-100-102,105), 2 IP (650-20,23), 1 6, 6 P, 3L (JA 650), 1 IP (651-34), 2 Ip? (652-10,12), 4 pd
(652-100,102,103,110), 4 p? (652-101,104,108,111), 2 IP (652-14,15), 11 L, 11(JA 652) [UCLA].
Parish not specified: 1944, R.B. Hill, 1 6; 1 Jan 1946, G.A. Thompson, 7 L; M. Grabham, 1 Ip;
G.A. Thompson, 1 6, 2 9? [USNM]; W.A. Page, 2 pd (771-101,106), 4 L (JA 771), 2 Ip? (772-33,
34), 8 pd (772-90,93,95,97,99,106,110,114), 11 p? (772-91,92,96,98,101 ,103,104,107,108,111,
113), 1 lp (772-30), 3 6,6 9,5 P, 11 p, 40 L, 11(JA 772) [UCLA].

10. Aedes (Howardina) inaequalis (Grabham)
Figs.3,7,8,23,24

1907.Howardina inaequalis Grabham, 1907:25-26. TYPE: Lectotype 6 with genitalia slide, Kings-
ton (St. Andrew), Jamaica, date not specified, M. Grabham [USNM; selection of Stone and
Knight 1956:219].

Aedes (Howardina) inaequalis in part of Thompson (1947:78; 1956:199-200); Stone, Knight and
Starcke (1959:174); Belkin, Schick and Heinemann (1965:31); Forattini (1965:399,400); Por-
ter (1967:38).

Aedes inaequalis in part of Stone and Knight (1956:219).

Howardina inaequalis of Theobald (1910:611); Surcouf and Gonzalez-Rincones (1911:141).

Aedes (Howardina) aurites in part of Dyar (1928:236); Edwards (1932:155); Lane (1939:99;
1953:720-722); Horsfall (1955:471).

Aedes aurites in part of Hill and Hill (1948:29,42).

Aedes (Howardina) aureostriata in part of Bonne and Bonne-Wepster (1925:370-371). 7

Aedes aureostriata in part of Howard, Dyar and Knab (1917:855-859); Dyar (1918:73,80).

FEMALE. Wing: 2.37 mm. Proboscis: 1.93 mm. Forefemur: 1.49 mm. Abdomen:
about 2.1 mm. Similar to aurites, differing primarily in the following. Head: Erect
scales dark. Thorax (fig.8): Mesonotal dark scales not dense as in aurites; mesonotal
light lines narrow, faintly golden, distinctly narrower than dark lines; outer dorsocen-
tral line shorter, not merging with lateral marginal line anteriorly. Lateral marginal
line narrow from side of anterior promontory to wingbase. Antealar area above para-
tergite with scattered dark scales only. Lateral scutellar lobes with a few narrow gold-
en scales. Apn with a restricted silvery patch, distinctly smaller than in aurites and

Berlin: Subgenus Howardina 47

grabhami; ssp with a distinct silvery patch; patches on base of pra, upper and lower
stp distinct, separate. Legs: Midtarsus with a few white scales on segments 1,2; hind-
tarsus with broader white rings on 1-3. Haltere: Stem pale, knob predominantly
white, with a few dark scales dorsally. Abdomen: Basomedian tergal straw yellow
patches on V-VII. |

MALE. Wing: 2.36 mm. Proboscis: 2.01 mm. Forefemur: 1.58 mm. Essentially
similar to female in coloration.

MALE GENITALIA (fig.23). Diagnostic characters as in the key. Ninth tergite lobe
not prominent, with 2,3 bristles. Sidepiece length about 2.5 median width. Clasper
length about 0.55 of sidepiece: spiniform short, about 0.25 of clasper.

PUPA (fig.23). Abdomen: 3.08 mm. Trumpet: 0.46 mm; index 5.0-5.6. Paddic:
0.68 mm. Chaetotaxy as given below, hairs lightly to moderately pigmented. Very
similar to stenei but apparently distinguishable by the moderately pigmented abdom-
inal tergite II and segment III only slightly darkened in the middle, without distinct
dark stripe. Readily separated from aurites and grabhami by the lightly pigmented
area on the metanotum. Cephalothorax: Mesonotum moderately pigmented between
trumpets; lateral area slightly darker; metanotum with a small to moderate, lightly
pigmented area in the middle of lateral portion. Hair 1(1), 2(2, 1-3), 3(1, 1-3), 4(4,
2-7), 5(6, 3-7), 6(1, 1-2), 7(4, 3-6), 8(1, 1-2), 9(2, 2-3), 10(15, 6-15), 111), 12(4, 2-
5). Abdomen: Tergite I moderately pigmented except for usual membranous areas;
tergites II,III moderately pigmented; other segments lighter. Segment I: hair 1(11-14
primary branches, subequal in length, each with many secondary branches), 2(1, 1-2),
3(1), 405, 4-7), 5(5, 4-6), 6(1), 7(3, 2-3), 9(1), 10(1, present in some), 11(1, present
in some). Segment II: hair 0(1), 1(40, 36-56), 2(1), 3(1), 4(4, 4-8), 5(1), 6(1), 7(4,
3-4), 9(1), 10(2, 2-3, present in a few). Segment III: hair 0(1), 109, 9-20), 2(1), 31),
4(4, 2-4), 5(1), 6(1), 7(4, 2-5), 8(3, 2-3), 9(1), 10(2), 11(1), 141). Segment IV: hair
O(1), 1(6, 3-14), 2(1), 3(4, 3-7), 4(2, 2-3), 5C1, 1-2), 6(1), 7(3, 2-4), 8@2, 1-2), 9C1),
10(2, 1-2), 11€1), 14(1). Segment V: hair 0(1), 1(1, 1-4), 2(1), 3@, 1-3), 4(6, 3-7),
S(1, 1-2), 6(1), 765, 3-7), 8@, 1-3), 9(1), 10(1, 1-2), 11€1), 141). Segment VI: hair
O(1), 1€1, 1-3), 2(1), 31, 1-3), 4(3, 2-7), 5C1, 1-2), 6(1, 1-2), 7(1, 1-2), 8@3, 2-4),
9(1), 10(€1), 111), 14(1). Segment VII: hair 0(1), 1(1), 21), 31), 4@, 1-3), 5(1),
6(3, 2-4), 7(1, 1-3), 8(3, 1-5), 9(4, 3-5), 10(€1), 11(1), 14(1). Segment VIII: hair 0(1),
4(2, 1-2), 9, 8-12), 14(1, 1-2). Paddle: Uniformly lightly pigmented, midrib slight-
ly darker; external margin smooth; hair 1-P slender. Male genital lobe extending to
0.56 of paddle, female genital lobe to 0.36.

LARVA (fig.24). Head: 0.7 mm. Siphon: 0.9 mm. Anal Saddle: 0.22 mm. Readily
separated from other species by the short, broad unequal gills and hair 5-C usually
double. Head: Width about 1.12 of length. Mental plate with a strong median tooth
and 11 distinct teeth (11-12) on either side, often a smaller detached tooth at base.
Hair 0(1), 1(1), 3(1), 4(3, 2-4), 5@2, 1-2), 6(9, 8-11), 7(4, 3-4), 8(1), 9(2, 2-3), 10(1),
— 1102, 2-5), 12(1, 1-2), 131), 14(1), 153, 2-3). Antenna: Length about 0.46 of head;
pigmentation very light; shaft with very few minute scattered spicules. Hair 1-A
about 0.5 from base. Thorax: Stellate hairs with a few unequal branches; long hairs
barbed and with attenuate apices. Prothorax: hair 0(4, 2-8), 1(3, 3-7), 2(1), 3(5, 4
6), 4€1, 1-2), 5(2), 6(1), 701, 1-2), 8(3, 2-4), 9(1, 1-2), 10(1), 11(2, 2-4), 12Q),
14(10, 9-12). Mesothorax: hair 1(2, 2-5), 2(1), 3(1), 4(1), 5(1), 6(4, 4-5),-7(1), 8(6,
5-8), 9(7, 5-8), 10(1), 11@2, 2-3), 12(1), 13(11, 7-11), 14(5, 4-5). Metathorax: hair
1(2, 2-4), 2(1), 3(4, 3-5), 4(2, 2-4), 5(5, 4-6), 6(1), 7(6, 5-7), 8G, 5-9), 9G, 5-7),
10(1), 11(2, 1-2), 12(1), 13(4, 1-5). Abdomen: Long and stellate hairs as on thorax.
Segment I: hair 1(3, 2-7), 2(5), 3(1), 4(3, 3-4), 5(4, 3-6), 6(3, 2-3), 7(1, 1-2), 9(1),

48 : Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

10(1), 11(6, 5-8), 13(3, 2-7). Segment II: hair 0(1), 1(2, 2-6), 2(5, 4-6), 3(1), 4@, 2-
A), 5(6, 4-8), 6(3, 2-3), 7(3, 2-4), 81, 1-2), 9(5, 4-6), 101), 11(1), 121), 13@, 2-4).
Segment III: hair 0(1), 1(2, 2-4), 2(5, 5-6), 3(1), 4(2), 5(5, 5-8), 6(1), 765, 3-6), 8(1),
9(5, 4-5), 10(1), 11(1), 12(1), 13(4, 2-8), 14(1). Segment IV: hair 0(1), 1@, 2-4),
2(6, 5-6), 3(1), 4(2), 5(7, 5-9), 6(1), 7(5, 3-6), 8(1), 9(5, 4-5), 10(1), 11€1), 121),
13(5, 3-10), 14(1). Segment V: hair 0(1), 1(3, 2-4), 2(5, 5-6), 3(1), 4(@2, 1-3), 5(6, 5-
9), 6(1), 7(5, 3-5), 8C1, 1-2), 965, 4-6), 10(1), 111), 12(1), 13(4, 2-9), 14C1). Seg-
ment VI: hair 0(1), 1(2, 2-3), 2(4, 4-5), 3(1), 4(1), 5(4, 2-6), 6(1), 7(3, 2-4), 8(3, I-
3), 9(3, 2-4), 10(1), 11(1), 121), 13(3, 24), 14(1). Segment VII: hair 0(1), 1(2),
I(1), 31), 401, 5C, 2-3), 64, 2-5), 70). 8C7; 6-9); 9G,.24), 1001), 1 LG), IP2c1),
13(2), 14(1). Segment VIII: Comb scales 16(14-18) in number; individual scale mod-
erately pigmented, elongate, fringed on sides and apex. Hair 0(1), 1(4, 3-5), 2(),
3(5, 5-6), 4(1), 5(3, 3-5), 14(1). Siphon: Index about 4.25 to 5.3; moderately pig-
mented but with darker basal ring. Pecten extending to about 0.4, 19(16-25) teeth;
individual tooth narrow, long, with minute spinules in distal half. Hair 1(2), 2(1), 3-5
not visible, 6(1), 7(1), 8(1), 9(1). Anal Segment: Saddle moderately pigmented. Gills
as long as or slightly shorter than saddle length, broad, unequal, ventral 0.64 of dor-
sal. Hair 1(1), 2(5, 4-5), 3(1), 4a(4, 3-4), 4b(4), 4c(4, 3-4), 4d(4), 4e(3), 4£(3, 1-3).

SYSTEMATICS. Contrary to the opinion of Thompson (1956:199), inaequalis is
a species distinct from aureostriatus (now grabhami) and aurites. The adults of these
3 forms are readily distinguished from one another by the characters mentioned in
the diagnoses and keys but I have not been able to find any reliable characters to sep-
arate inaequalis from Thompson’s stenei which, however, I consider to be also a dis-
tinct species. The larva of inaequalis is clearly differentiated from the other members
of the group but the pupa is very similar to that of stenei. Furthermore, an ecological
barrier separates inaequalis from the rest of the group as this species is never found
breeding in bromeliads to which the other species are confined. A. inaequalis is found
primarily at low elevations and seldom above 1,500 feet. I have seen specimens of in-
aequalis only from the parishes of Clarendon, Kingston, Manchester, Portland, St.
Andrew, St. Catherine, St. Mary and St. Thomas. This species may be more widely
distributed and more common than the records indicate because these are based on
immature stages only and these breeding sites have not been as widely sampled as
those of the other species of the group. :

BIONOMICS. A. inaequalis has been found breeding primarily in treeholes and
cut and broken bamboo but has also been taken occasionally in rockholes and crab-
holes. There are no records of this species breeding in bromeliads. Since adults cannot
be distinguished at present from stenei, it is not known if females of inaequalis bite
humans but it is possible that some females of this species are included in the stenei-
inaequalis material obtained in the 24-hour biting-landing survey (fig.7).

DISTRIBUTION (fig.3). Endemic to the island of Jamaica where it has been re-
ported from the parishes of Clarendon, Kingston, Manchester, Portland, St. Andrew,
St. Catherine, St. Mary and St. Thomas from near sea level to about 1000 ft. Mate-
rial examined: 574 specimens; 55 6, 81 2, 99 pupae, 339 larvae; 83 individual rear-
ings (47 pupal, 28 larval, 8 incomplete).

JAMAICA. Clarendon: May Pen, 10 Sept 1965, rockholes, J.N. Belkin (JA 352), 6 pd (352-
100A,102,104,106,107,109), 6 p? (352-32,90,91 93,108,111), 1 IP (352-12), 6 6, 8 2, 1 P, 8 p,
37 L; 6 Sept 1966, rockholes, D.C. Watson (JA 602), 1 pd (602-101), 1 p? (602-100), 5 6, 7 9;
1 Sept 1966, rockholes, D.C. Watson, 2 6 (JA 603) [UCLA] . Manchester: Hope Farm, elev. 900-
1000 ft, 23 Oct 1966, treehole, O.G.W. Berlin and D.C. Watson (JA 634), 2 Ipd (634-10,17), 5 Ip? —
(634-11-13,15,19), 2 pd (634-102,104), 3 p? (634-100-102), 4 IP (634-14,16,18,20), 1 L, 11

Berlin: Subgenus Howardina 49

[UCLA]. Portland: Kemney-Sherwood Forest, elev. 750 ft, 13 Aug 1967, bamboo, J.N. Belkin
and W.A. Page, 1 lpd (713-10), 1 pd (713-100), 16 L(JA 713), 2 Ipd (715-10,11), 2 pd (715-100,
102) [UCLA]. St. Andrew and Kingston: Constant Springs, 16 Nov 1966, treehole, O.G.W. Berlin
and D.C. Watson, 6 L (JA 696) [UCLA]. Ferry Quarry, 13 Sept 1965, crabhole, W.A. Page, 1 ?
(JA 369) [UCLA]. Hermitage Dam Road, 20 Aug 1964, rockhole, L. Paton (JA 109), 2 Ipd (109-
10,11), 1 pd (109-101), 1 p? (109-100), 4 3, 1°,1P,2L;22 Aug 1965, treehole, W.A. Page (JA
280), 1 pd (280-102), 1 6; 10 Aug 1965, treehole, W.A. Page (JA 288), 2 Ip? (288-10,15), 1 pd
(288-102), 1 6; 13 Aug 1965, bamboo, W.A. Page, 1 Ipd (290-14), 3 Ip? (290-10,11,13),2L(JA
292), 25 L (JA 293), 6 L(JA 294), 1 Ip? (295-23), 26, 19,10 L(JA 296), 1 L(JA 298),12(JA
299), 1 p? (300-20), 2 L (JA 300), 17 L (JA 302), 1 lp? (304-22), 3 L (JA 307), 24 L(JA 308),
1d, 19, 1 P (JA 309); 7 Sept 1965, J.N. Belkin, 1 9, 2 L(JA 330), 3 L(JA 331); 29 Oct 1965,
treehole, W.A. Page, 1 pd (390-100A), 2 p? (390-101,102), 1 p? (397-105); 13 July 1967, tree-
hole, W.A. Page (JA 826), 1 p? (826-20) [UCLA]. Mt. Salus, 17 Sept 1967, bamboo, W.A. Page,
123 L (JA 780) [UCLA]. Temple Hall, elev. 500 ft, 20 Aug 1965, bamboo, W.A. Page, 1 p? (250-
102), 1 6,1 9,1 P,5 L(JA 250), 1 Ipd (252-35), 4 Ip? (252-102,105,107,108), 16,1 p,6L,11
(JA 252); 25 Aug 1965, bamboo, D.C. Watson (JA 251), 1 pé (251-107), 1 IP (251-60), 1,11 L
[UCLA] . Waterworks, 7 Sept 1965, treehole, J.N. Belkin (JA 327), 1 p? (327-104), 19,1 p,1L;
3 Dec 1965, treehole, W.A. Page (JA 408), 1 IP (408-14) [UCLA]. St. Catherine: Harkers Hall,
elev. 750 ft, 12 July 1967, bamboo, W.A. Page (JA 822), 2 lp? (822-12,14), 1 IP (822-10), 1 ,
1 p, 5 L [UCLA]. Rio Cobre Dam, elev. 100 ft, 10 Sept 1965, treehole, J.N. Belkin (JA 351), 1 lp?
(351-30), 1 p? (351-105), 1 IP (351-31), 1 L; 17 Sept 1965, ditch, D.C. Watson, 1 6 (JA 372), tree-
hole, 1 6,3 2, 1 p(JA 373) [UCLA]. Spanish Town, 9 Sept 1965, treehole, D.C. Watson (JA 605),
1 pd (605-102) [UCLA]. St. Mary: Prospect, 3 Nov 1964, rockhole, W.A. Page, 1 6,19,1p(JA
173) [UCLA]. St. Thomas: Folly, elev. 50 ft, rockhole (JA 103), 4 p? (103-102,103,105,108), 1?
[UCLA] . Parish not specified: Jan 1940, bamboo, 1 6, 2 ? [USNM].

11. Aedes (Howardina) stenei Thompson
Figs 3,7 .6.22,20

1956.Aedes stenei Thompson, 1956:196-199. TYPE: Holotype 6 (45090211) with associated lar-
val skin and male genitalia slide, near Hope Farm, just above Porus, on road to Mandeville,
Manchester (as Clarendon), Jamaica, elev. about 1000 ft, Sept 1945, G.A. Thompson
[USNM].

Aedes (Howardina) stenei of Stone, Knight and Starcke (1959:174); Belkin, Schick and Heine-
mann (1965:31); Forattini (1965 :471); Porter (1967:38).

FEMALE. Wing: 2.57 mm. Proboscis: 1.66 mm. Forefemur: 1.58 mm. Abdomen:
about 2.02 mm. Essentially similar to inaequalis from which it cannot be separated
(fig.8). |
_ MALE. Wing: 2.38 mm. Proboscis: 2.1 mm. Forefemur: 1.58 mm. Essentially sim-
ilar to female in coloration.

MALE GENITALIA (fig.25). Diagnostic characters as in the key. Ninth tergite lobe
indistinct, with 2,3 strong bristles. Sidepiece length about 2.5 median width. Clasper
about 0.6 of sidepiece length, basal 0.33 with small spicules; spiniform about 0.25 of
clasper.

PUPA (fig.25). Abdomen: 2.24 mm. Trumpet: 0.38 mm; index about 5.0. Paddle:
0.68 mm. Chaetotaxy as given below, hairs lightly to moderately pigmented. Very
similar to inaequalis but distinguished by the deeply pigmented abdominal tergite II
and by the presence of a distinct narrow median longitudinal dark stripe on tergite
III. Readily separated from aurites and grabhami by the 2-branched hair 9-C and the
lightly pigmented area on the metanotum. Cephalothorax: Mesonotum moderately

50 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

pigmented between trumpets, lateral areas slightly darker; metanotum with a lightly
pigmented area in the middle of lateral portion. Hair 1(1), 2(2, 2-3), 3(1), 405, 4-7),
5(5, 4-9), 6(1, 1-2), 7(4, 3-9), 8(1, 1-2), 9@2, 2-3), 10(10, 7-18), 11(1), 12(5, 3-7).
Abdomen: Tergite I moderately pigmented except for usual membranous areas; ter-
gite II strongly pigmented; tergite III with distinct narrow, median, dark longitudinal
stripe; pigmentation progressively lighter caudad. Segment I: hair 1(12-17 primary
branches, subequal in length each with several secondary branches), 2(1, 1-2), 3(1),
4(6, 4-8), 5(4, 3-8), 6(1), 7(3, 1-3). 9(1), 10(1), 11(1). Segment IT: hair 0(1), 1(46,
36-72), 2(1), 3(1), 4(5, 3-6), 5(1), 6(1), 7(4, 2-5), 9(1), 10€1, 1-4, present in some).
Segment III: hair 0(1), 1(16, 7-23), 2(1), 3(1, 1-2), 4(3, 2-4), 5(1, 1-2), 6(1), 7(3, 2-
4), 8(2, 2-4), 9(1), 10(2, 1-2), 111), 14(1). Segment IV: hair 0(1), 1(10, 4-19), 21),
3(5, 3-6), 4(3, 1-4), 5(1, 1-3), 6(1), 7(@2, 2-4), 8(2, 1-3), 9(1), 10@, 1-2), 11(1), 14(1).
Segment V: hair 0(1), 1(2, 1-7), 2(1), 3@, 1-3), 4(4, 3-6), 51, 1-2), 6(1), 7(4, 2-5),
8(2, 1-3), 9(1), 10(1, 1-2), 11€1), 14(1). Segment VI: hair 0(1), 1(2, 1-4), 21), 3@,
1-2), 4(3, 2-5), 5(1, 1-4), 6(1, 1-2), 7(1), 8(2, 2-4), 9(1), 10(€1), 11€1), 14(1). Seg-
ment VII: hair 0(1), 1(1), 21), 31, 1-2), 4(2, 1-2), 5(1), 6(3, 2-4), 7(1), 8(4, 3-5),
9(5, 3-6), 10(1), 11€1), 14(1). Segment VIII: hair 0(1), 42, 1-2), 9(10, 8-13), 14(1).
Paddle: Uniformly very lightly pigmented, midrib slightly darker; external margin
smooth. Male genital lobe extending to 0.42 of paddle, female genital lobe to 0.38.
LARVA (fig.26). Head: 0.7 mm. Siphon: 0.95 mm. Anal Saddle: 0.27 mm. Read-
ily separated from other members of the group by the strong, 4-7-branched hair 2-
VII. Head: Width about 1.02 of length. Mental plate with a strong median tooth and
11 distinct teeth (9-11) on each side, often a detached tooth at base. Hair 0(1), 1(1),
3(1), 4(5, 3-5), 5(4, 3-5), 614, 11-14), 7(5, 4-7), 8(1), 9(2, 2-3), 10(1), 11(6, 3-8),
12(2, 1-2), 13(1), 141), 15(2, 2-3). Antenna: Length about 0.45 of head; pigmenta-
tion very light; shaft with a few scattered spicules. Hair 1-A about 0.5 from base.
Thorax: Stellate hairs with a few, unequal, weakly barbed branches; long hairs barbed
and with attenuate apices. Prothorax: hair 0(12, 10-14), 1(9, 7-11), 2(1), 3(14, 8-
16), 4€1, 1-2), 5(2, 2-3), 6(1), 7(2, 2-3), 8(6, 6-10), 9(1, 1-2), 10(1), 11(3, 3-5),
12(1), 14(10, 8-14). Mesothorax: hair 1(5, 3-9), 2(1), 3(1), 4(1), 5(1), 6(5, 4-5),
7(1), 8(6, 5-7), 9(8, 7-10), 10(1), 11(2, 1-2), 12(1), 13(14, 10-14), 14(4, 4-9). Meta- -
thorax: hair 1(2, 2-8), 2(1), 3(3, 3-5), 4(8, 5-9), 5(8, 6-11), 6(1), 7(7, 7-9), 8(6, 4-9),
9(8, 6-8), 10(1), 11@2, 1-3), 12(1), 13(11, 5-13). Abdomen: Stellate hairs as on tho-
rax. Segment I: hair 1(5, 5-12), 2(8, 5-9), 3(1), 4(3, 2-4), 5(8, 7-11), 6(3), 7(1), 9(1),
10(1), 11(8, 7-9), 13(10, 3-12). Segment II: hair 0(1), 1(3, 2-9), 2(7, 3-8), 3(1), 4(3,
2-3), 5(9, 7-11), 6(3), 7(2, 2-3), 8(1, 1-2), 9(6, 5-7), 10(1), 11(1), 12(1), 13(8, 5-12).
Segment III: hair 0(1), 1(3, 2-5), 2(7, 5-8), 3(1), 4(2), 5(10, 9-11), 6(1, 1-2), 7(6, 4-
7), 8(1), 9(6, 5-7), 101), 1101), 121), 1309, 5-13), 14(1). Segment IV: hair 0(1), 103,
2-4), 2(7, 6-8), 3(1), 4(2), 5(9, 9-12), 6(1, 1-2), 7(5, 4-6), 8(1), 9(6, 5-6), 10(1),
11(1), 12(1), 13(10, 7-11), 14(1). Segment V: hair 0(1), 1(2, 2-3), 2(6, 5-7), 3(1),
4(3, 2-4), 510, 9-11), 6C1, 1-2), 7(5, 4-6), 8(1), 9(6, 5-6), 10(1), 111), 12(1), 13(8,
4-11), 14(1). Segment VI: hair 0(1), 1(2, 2-3), 2(6, 5-7), 3(1), 4(1, 1-2), 5(7, 5-10),
6(1), 7(2, 2-3), 8(3, 2-3), 9(5, 4-6), 10(1), 11(1), 12(1), 1309, 6-11), 14(1). Segment
VII: hair O(1), 1(2, 2-4), 2(5, 4-7), 3(1), 4(1), 5(2, 2-3), 6(2, 2-3), 7(1), 8(5, 3-7),
9(4, 4-6), 10(1), 11(1), 121), 13(6, 3-6), 14(1). Segment VIII: Comb scales 18(16-
19) in number, individual scale moderately pigmented, elongate and fringed on sides
and apex. Hair 0(1), 1(4, 4-6), 2(1), 3(6, 5-7), 4(1), 5(4, 3-5), 14(1). Siphon: Index
about 5.8-6.7; moderately pigmented, with darker basal ring. Pecten extending to
about 0.43; 19(19-24) narrow, long teeth. Hair 1(2, 2-3), 2(1), 3-5 not visible, 6(1),
7(1), 8(2, 1-2), 9(1). Anal Segment: Saddle moderately pigmented. Gills subequal;

Berlin: Subgenus Howardina 51

moderate, less than 2.0 of saddle length. Hair 1(1, 1-2), 2(7, 5-8), 3(1), 4a(4, 4-5),
4b(4, 3-4), 4c(4, 3-4), 4d(4), 4e(4, 3-4), 4£(3, 3-4).

SYSTEMATICS. A. stenei appears to be most closely related to inaequalis. | have
not found any reliable characters to separate the adults of these 2 species and their
pupae are also very similar but apparently differentiated by the features mentioned
in the diagnoses and key. The larva of stenei is readily separated from inaequalis and
the other members of the group. There appears to be a strong ecological barrier be-
tween stenei, which breeds exclusively in bromeliads, and inaequalis, which is pri-
marily a treehole breeder and has never been found in bromeliads. A. stenei has been
collected most frequently at elevations of about 1,000 feet and less commonly down
to 100 feet and as high as 2,500 feet. Its most common associate is grabhami; as in-
dicated under the latter there is a suggestion of occasional hybridization between
these 2 forms. A. stenei is the most common and widespread species of the group
and has been reported from all the parishes of Jamaica except Clarendon, Trelawny
and Westmoreland.

BIONOMICS. A. stenei has been found breeding only in terrestrial and epiphytic
bromeliads. Although the females of stenei and inaequalis cannot be separated with
certainty, I believe that the majority of the stenei-inaequalis specimens collected in
the 24-hour biting-landing study reported in fig.7 were probably stenei. These showed
a small peak of activity at 1700 hrs.

DISTRIBUTION (fig.3). Endemic to the island of Jamaica where it has been re-
ported from all the parishes except Clarendon, Trelawny and Westmoreland at eleva-
tions of 100 to 2500 ft. Material examined: 1486 specimens; 226 6, 380 2, 492 pu-
pae, 388 larvae; 371 individual rearings (227 pupal, 102 larval, 42 incomplete).

JAMAICA. Hanover: Lethe, elev. 500 ft, 6 Nov 1966, O.G.W. Berlin and D.C. Watson (JA 666),
2 p? (666-100,101), 2 p, 2 L [UCLA]. Manchester: Hope Farm, 2 Sept 1945, G.A. Thompson,
1 Ip? (450902-2); 2 Oct 1945, G.A. Thompson, 2 Ip? (451002-10,13), 1 lp? paratype (451002-3),
1 Ipd, paratype (451002-2); 1 Jan 1946, G.A. Thompson, 1 Ipd, paratype (460101-20), 1 lp?
(460101-18) [USNM]; elev. 900-1000 ft, 18 Aug 1965, W.A. Page, 1 lpd (240-62), 2 Ip? (240-63,
64), 1 IP (240-65), 1 Ip? (241-62), 7 pd (241-101,103,106-108,110,111), 1 p? (241-114), 1 IP
(241-60), 2 pd (242-62,100), 4 p? (242-61,103,105,106); 10 Sept 1965, W.A. Page, 1 lp? (355-
10), 1 p? (355-100), 2 6, 2 2 (JA 355), 2 2 (JA 356); 2 Nov 1966, O.G.W. Berlin, 3 Ipd (661-10,
11,16), 2 pd (661-100,104), 3 p? (661-101-103), 4 IP (661-12-15), 19,2 p,61(JA 661), 3 2(JA
663A); 12 Nov 1966, 0.G.W. Berlin, 90 ? (JA 684), 3 P, 5 L(JA 685), 3 lpdé (688-14,17,19), 2 lp?
(688-10,18), 9 pd (688-105,106,112,117-119,124,125,127), 1 p? (688-109), 5°, 1 p,91(JA 688)
[UCLA]. Porus, elev. 900-1000 ft, 23 Oct 1966, O.G.W. Berlin and D.C. Watson, 1 lp? (630-12),
1 IP (630-13), 1 6, 1 p, 3 L(JA 630), 2 Ipd (631-13,14), 4 lp? (631-10,11,16,17), 4 pd (631-100,
101,104,107), 6 p?(631-102,103,110,111,114,115), 5 IP (631-18-22), 2d, 1 9, 1 p(JA 631), 1 Ipd
(632-14), 3 pd (632-109-111), 1 p? (632-113), 1 IP (632-16), 1 6, 19,2P,2 p, 5 L, 31; 1 Nov
1966, O.G.W. Berlin, 1 p? (655-102), 1 1(JA 655), 3 lp? (658-10,11,13), 6 pd (658-100-102,106,
109,111), 3 p? (658-104,105,108), 1 IP (658-12), 1 9, 2 p, 21 [UCLA]. Portland: Kemney-Sher-
wood Forest, elev. 750 ft, 13 Aug 1967, J.N. Belkin and W.A. Page, 2 p? (706-12,103), 146,19
(JA 706), 1 Ip? (708-10), 1 6, 1 9, 2 p, 21(JA 708), 2 L(JA 710), 2 p(JA 711) [UCLA]. St. An-
drew and Kingston: Hermitage Dam Road, 1 Sept 1965, W.A. Page (JA 319), 1 pd (319-102); 29
Oct 1965, W.A. Page (JA 392), 3 pd (392-103,107,112), 2 p? (392-108,110); 15 Aug 1967, JN.
Belkin and W.A. Page, 1 lp? (723-25), 2 p? (723-100,101), 1 IP (723-22), 1 9, 2P (JA 723), 2 Ipd
(725-22,23), 2 Ip? (725-11,12), 7 pd (725-94,101,102A,103A,108,111,112A), 9 p? (725-100,103,
105B,106A,107A,110,111A,113,114), 2 IP (725-20,21), 3 ?, 1 P, 5 p (JA 725); 25 July 1967,
W.A. Page, 3 pd (854-100,101,103), 1 p? (854-104), 1 Ip (854-12), 1 P, 11(JA 854), 1 P(JA 855)
[UCLA]. Kingston, 15 Sept 1906, M. Grabham, 1 6 [USNM]. Mavis Bank, M. Grabham, 1 6
[USNM]. Mt. Salus Road, 11 Sept 1967, W.A. Page (JA 779), 2 pd (779-102,106), 1 p? (779-103)
[UCLA] . Newcastle, elev. 2500 ft, 29 Oct 1966, O.G.W. Berlin, 3 L (JA 648) [UCLA]. Shooters

oo Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Hill, elev. 1800 ft, 3 Dec 1946, G.A. Thompson, 2 ¢ [IJ]. Stony Hill, 16 Sept 1946, G.A. Thomp-
son, 1 ? [IJ]. Temple Hall, elev.500 ft, 27 Oct 1966, O.G.W. Berlin and D.C. Watson, 3 L (JA 646)
[UCLA]. St. Ann: Moneague, 1943, R.B. Hill, 1 9 [USNM]; 9 Sept 1966, D.C. Watson, 1 d,1¢
(JA 607); 17 Aug 1967, J.N. Belkin and W.A. Page (JA 733), 2 lpd (733-10,15), 9 Ip? (733-11-14,
16-19,30), 7 pd (733-90,94,97,104,107,110,113), 12 p? (733-31 ,41,92,93 ,99-102,106,111,112),
12 6,19 9,2 P, 18 p, 11 L, 31; 6 Aug 1967, W.A. Page (JA 771), 1 p? (771-102), 1 L [UCLA] ;
Faiths Pen Schoolyard, elev. 1750 ft, 30 Aug 1967, W.A. Page (JA 762), 1 Ipd (762-11), 2 Ip?
(762-12,13), 1 19 (762-14), 1 pd (762-103), 1 p? (762-102), 1 P, 9 L, 1 1 [UCLA]; 2.0 mi S of
Moneague near Unity, elev. 1750 ft, 26 Aug 1967, W.A. Page (JA 758), 1 lp? (758-23), 4 p? (758-
31,32,108,111), 5 p? (758-105-107,112,114), 8 6, 6 2,5 p, 15 L [UCLA] ; 29.0 mi post on Mon-
eague Road, elev. 1750 ft, 30 Aug 1967, W.A. Page, 1 Ipd (763-20), 6 Ip? (763-12,14,21-23 30),
8 pd (763-90,94,96-98 100,107,111), 6 p? (763-95 ,99,101-103,105), 2 6,12 9,11 P, 19 p, 18 L,
1 1(JA 763), 1 lpd (764-31), 1 Ip? (764-34), 5 pd (764-32,96,112-114), 4 p? (764-90,93,107,111),
1 lp (764-33),6 6,8 9,5 P,5 p,1L,11 [UCLA]; Moneague-Claremont Road, 17 Sept 1965, W.A.
Page, 1 pd (380-106), 1 p (JA 380), 1 Ip? (381-12), 1 pd (881-100) [UCLA]; Moneague-Ocho Rios
Road, 10 Sept 1967, W.A. Page (JA 776), 1 pd (776-100) [UCLA]. Mt. Diablo, 17 Sept 1965,
W.A. Page, 2 Ipd (375-11,14), 2 p? (375-10,100), 2 lp (375-1522), 8 9, 21(JA 375), 3 Ipd (376-
10,11,13), 2 lp? (376-15,16), 2 pd (376-100,102), 3 P (JA 376); 17 Aug 1967, J.N. Belkin and
W.A. Page, 1 Ipd (731-22), 2 Ip? (731-20,23), 1 pd (731-96), 1 p? (731-101), 16,2?,2p,3 LGA
731), 2 lp? (732-30,31), 4 pd (732-95,103,109,110), 2 p? (732-32,98), 1 IP (732-33), 2 P, 2 p, 7 L,
11(JA 732), 2 lp? (734-10,11), 4 pd (734-103 ,104,106,109), 3 p? (734-100,107,108), 1 IP (734-
12),2 9,1 p, 22 L(JA 734); 7 Sept 1967, W.A. Page, 1 lp? (779-10), 1 p? (774-107), 4 L (JA 774)
_ [UCLA]. St. Catherine: Bog Walk, elev. 500 ft, 12 July 1967, W.A. Page (JA 820), 1 pd (820-11)
[UCLA]. St. Elizabeth: Maggotty, elev. 500 ft, 15 Sept 1967, W.A. Page (JA 791), 1 L [UCLA].
St. James: Five Miles, elev. 100 ft, 6 Nov 1966, O.G.W. Berlin and D.C. Watson (JA 676), 1 lp?
(676-12), 4 pd (676-102,103,105,124), 8 p? (676-100,101,106,108,109,117,122,125), 4 d, 2 @,
2 L [UCLA]. Montego Bay, elev. 500 ft, 6 Nov 1966, O.G.W. Berlin and D.C. Watson (JA 669),
1 lpd (669-10), 1 lp? (669-11), 2 pd (669-100,102), 1 p? (669-101), 3 L, 11 [UCLA]; 16 Jan 1940,
26,2 2;17 Jan 1940, 2 6, 3 9? [USNM]. Reading, elev. 200 ft, 6 Nov 1966, O.G.W. Berlin and
D.C. Watson, 3 pé (670-100-102), 1 p? (670-103), 3 P, 3 p, 5 1(JA 670), 2 lp? (671-10,14), 1 pd
(671-100), 4 IP (671-11-13,15), 13 L (JA 671), 1 pd (673-101), 1 p? (673-100), 1 IP (674-10), 1 p,
4 L (JA 674) [UCLA]. St. Mary: Broadgate, elev. 500 ft, 27 Oct 1966, O.G.W. Berlin and D.C.
Watson (JA 643), 4 Ipd (643-32,38-40), 3 lp? (643-34-36), 6 pd (643-104,106-108,112,113), 6 p?
(643-30,100-103,105), 2 IP (643-41,42), 3 6,2 %,2P,3 L,51 [UCLA]. St. Thomas: Port Morant
Bay, 16 Aug 1967, J.N. Belkin and W.A. Page (JA 729), 3 Ipé (729-10,11,20), 1 pd (729-102),
4 p? (729-100,101,103,104), 7 L, 21 [UCLA]. Windsor Forest, elev. 750 ft, 30 Oct 1966, O.G.W.
Berlin and D.C. Watson, 1 Ip? (650-24), 3 pd (650-104,106,107), 1 p? (650-108), 1 IP (650-20),
3 6,29,2L,21(JA 650), 1 Ipd (651-37), 2 lp? (651-3539), 4 pd (651-110-113), 1 p? (651-114),
5 IP (651-31-33,36,38), 1 p, 11 (JA 651), 2 Ipd (652-11,13), 2 pd (652-106,109), 2 p? (652-105,
107), 3 L, 21 (JA 652) [UCLA]. Parish not specified: New Park, June 1929, C.D. Williams, 1
[BM]. Apr 1906, M. Grabham, 1 2; 1944, R.B. Hill, 4 6, 1 9; Jan 1940, W.H.W. Komp, 2 6; R.B.
Hill, 1 6, 4 9? [USNM]; W.A. Page, 2 lpdé (771-10,12), 3 pd (771-100,104,107), 3 p? (771-13,102,
105), 1 P, 1 p, 28 L (JA 771), 2 Ip? (772-32,35), 2 pd (772-100,112), 2 p? (772-102,105), 2 d,
5 9,2P,3 p,15 L(JA 772) [UCLA].

Ioliota Group

FEMALES. Head: Narrow decumbent scales white. Palpus with apical silvery
white scales. Thorax: Inner dorsocentral line narrow, pale golden; median prescutel-
lar line broad, white; outer dorsocentral line narrow, pale golden; lateral marginal
line narrow, white, nearly continuous from anterior promontory to wingbase. Acros-
tichal bristles not developed on disc. Scutellum predominantly dark scaled, light

Berlin: Subgenus Howardina 53

scales restricted to median lobe. Ssp with distinct silvery patch; upper stp patch sep-
arate, small. Legs: Foretarsus with a few white scales on base of segment 2; midtar-
sus with small incomplete basal white rings on segments 1,2; hindtarsus with broader
complete basal white rings on segments 1-3 and with segment 5 completely or largely
white. Abdomen: Basomedian tergal white markings dingy white or creamy on seg-
ments III-VI or V-VI, white on VII.

MALES. Palpus with submedian white scales on segments 4,5. Larger claw of fore-
leg and midleg with small acute basal tooth and large blunt submedian tooth; smaller
Claw with basal tooth only.

MALE GENITALIA. Claspette with short broad conical tubercle bearing 1 or
rarely 2 apical bristles. Clasper less than 0.5 of sidepiece; with distinct subapical
hairs. Aedeagus subquadrate, shorter than proctiger.

PUPAE. Cephalothorax: Hair 5-C single or double, at least 2.0 of hair 4-C; hair
7-C usually single; hair 10-C single to triple. Abdomen: All large hairs except 1-IL III,
9-VII,VII usually single or double; hair 2-II laterad of or in line with hair 3; hair 5-
IV,V short to moderate, not exceeding middle of tergite following; hair 9-II-VI cau-
dad of hair 6; 9-VII longer, stouter and with more branches than hair 6-VII. Paddle:
Moderate; apex rounded; marginal spicules indistinct.

LARVAE. Head: Hair 1-C unbranched; 4-C moderately long, multiple; 5,6-C both
long, single; hairs 11,14-C weakly stellate, 11-C with unequal branches; 15-C double,
long, exceeding apex of mental plate. Antenna: Shaft with a few minute sharp spic-
ules; hair 1-A single. Thorax: Hair 13-P present. Hairs 0,8,13,14-P, 1,4,5,8,13-T
weakly stellate; hair 1-T far mesad of hairs 2,3. Abdomen: Integument glabrous dor-
sally, ventral surfaces of segments II-VIII with rows of minute spinelike spicules; hair
2 distinctly removed cephalad but not beyond hair 8, hair 9 not markedly displaced
cephalad. Hairs 2,5-I-VI, 9-II-VI, 11-I distinctly stellate, with a few branches, 2,9,11
distinctly heavier than others, hairs 1-I,II, 7-II-VI, 9-I very weakly stellate; hair 8-VII
distinctly cephalodorsad of 9-VII. Segment VIII: Comb scales in a single row; indi-
vidual scales moderate in length, blunt apically and finely fringed. Siphon: Index less
than 3.0; integument without spicules; distinct attached acus present. Pecten extend-
ing to about middle of siphon, not reaching hair 1-S; teeth numerous, closely set,
with basal denticles. Anal Segment: Saddle body without distinct spicules; caudal
margin with short pointed spines preceded by a few rows of spicules; hair 1-X usual-
ly single (1-3), near lower end of caudal margin. Ventral brush with 6 pairs of hairs
on a small sclerotized boss with distinct grid bars.

DISCUSSION. The Ioliota Group of 2 very similar species is most strikingly dif-
ferentiated from the other groups of the section by the presence of hair 13-P in the
larva, a character not found anywhere else in the section. In adult ornamentation
this group is very similar to the Albonotatus Group, differing from it primarily in the
presence of light scaling on hindtarsal segment 5.

The 2 species have widely disjunct distributions, cozumelensis occurring only on
the Yucatan Peninsula and the adjacent Cozumel island and British Honduras, and
ioliota at the opposite end of the Caribbean sea in northern Venezuela and Trinidad
(fig.3). The 2 species are so similar that they obviously belong to the same phylad.
Their pattern of distribution is interpreted here as a relict of a former wide distribu-
tion of the group in the central Caribbean area. A similar pattern is found in the
Walkeri Group.

A larva in a recent collection from a cut bamboo from the province of Guayas
(ECU 161) indicates the presence of another member of the Ioliota Group in Ecua-
dor. It differs from other members of the group by a 14-16 branched stellate hair

54 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

13-P which is only 3-5 branched in ioliota and double in cozumelensis. The albono-
tatus of Levi-Castillo (1952a:77; 1952b:555; 1953:40) is probably this member of
the Ioliota Group.

Both members of the group were probably originally treehole breeders. They have
been reported primarily from artificial containers and ioliota has been found breed-
ing in fallen cacao pods. There are no records of this group from bromeliad axils.

12. Aedes (Howardina) ioliota Dyar and Knab
Figs.3,8,27,28

1913.Aedes ioliota Dyar and Knab, 1913:76-78. TYPE: Holotype ?, Trinidad, locality not speci-
fied, F.W. Urich [USNM, 16000].

Aedes (Howardina) ioliota of Bonne and Bonne-Wepster (1925:371-372); Dyar (1928:233); Ed-
wards (1932:155); Lane (1939:100; 1953:719-720); Anduze (1941:14); Horsfall (1955:471);
Stone and Knight (1956:219-220); Stone, Knight and Starcke (1959:174); Aitken (1960:4);
Suarez and Cova Garcia (1961:17-24); Stone (1963:127); Belkin, Schick and Heinemann (1965:
69); Forattini (1965:401); Cova Garcia, Sutil and Rausseo (1966a:57,283; 1966b:73-74,141,
169,334).

Aedes ioliota of Urich (1913:528); Dyar (1918:73).

FEMALE. Wing: 3.33 mm. Proboscis: 2.62 mm. Forefemur: 2.45 mm. Abdomen:
about 2.98 mm. As described for the group, with following apparently diagnostic fea-
tures. Head: Median stripe on vertex narrow, white; light scales on sides and venter
straw yellow. Erect scales dark brown. Palpus about 0.16-of proboscis. Thorax (fig.
3): Scales on anterior promontory white. Inner dorsocentral line narrow, creamy,
reaching prescutellar space, followed by a moderate, white-scaled median prescutellar
line; outer dorsocentral line narrow, golden; lateral marginal line white, narrow, long,
extending from side of anterior promontory to wingbase, apparently interrupted on-
ly at scutal angle. Antealar area above paratergite with scattered dark scales. Light
scales on median scutellar lobe narrow, white; lateral lobes with a few narrow dark
scales. Pleural scales as in the section and group. Legs: Forefemur and midfemur pre-
dominantly dark anteriorly, a few basal scales creamy, proximal 0.5 of posterior sur-
face straw yellow, extending distally as a narrow line, rest dark; hindfemur predomi-
nantly straw yellow to 0.5 on both sides, rest dark scaled. Hindtarsal segment 5 com-
pletely white scaled. Haltere: Stem pale, knob mostly whitish, a few dark scales dor-
sally. Abdomen: Sternites predominantly creamy on II,III; rest with basal creamy
and apical dark scales, latter increasing in extent on distal segments.

MALE. Wing: 2.79 mm. Proboscis: 2.54 mm. Forefemur: 1.93 mm. Essentially
similar to female in coloration.

MALE GENITALIA (fig.27). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with 5,6 strong bristles. Sidepiece length about 2.5 median
width; basolateral scales white, remainder dark. Claspette with 2,3 small subapical
hairs. Clasper about 0.5 of sidepiece, with 1 subapical hair; minute spicules on basal
0.5. Dorsal sclerotization of aedeagus very shallowly emarginate, but without distinct
lateral lobes.

PUPA (fig.27). Abdomen: 3.35 mm. Trumpet: 0.43 mm; index 4.4-4.6. Paddle:
0.74 mm. Chaetotaxy as given below, hairs moderately to darkly pigmented. Diag-.
nostic characters as in the key. Apparently distinguished from cozumelensis by the

Berlin: Subgenus Howardina 55

more dendritic hair 1-IJ and longer, ovoid or slightly produced paddle, with length
about 1.5 greatest width. Cephalothorax: Mesonotum moderately pigmented between
trumpets, lateral areas and metanotum deeply pigmented; trumpets strongly con-
trasting. Hair 1(1), 2(2, 1-2), 3(1), 4(2, 1-3), 5(1, 1-2), 6(1), 7(1), 8(3, 2-4), 91),
10(2, 1-3), 111, 1-2), 12(1, 1-3). Abdomen: Tergite I moderately pigmented except
for usual membranous areas; segments II and III imbricate, moderately pigmented;
rest of segments progressively lighter caudad. Segment I: hair 1(6-12 primary branches,
each with many secondary ones), 2(1), 3(1), 4(3, 2-4), 5(2, 2-3), 6(1), 7(1), 9(1).
Segment II: hair 0(1), 1(3, strongly dendritic), 2(1, 1-2), 3(1, 1-2), 4(4, 2-5), 5(1),
6(1), 7(1), 9(1). Segment III: hair 0(1), 12, 1-4), 2(1), 31), 4@2, 1-4), 5(1), 6(1),
7(2, 1-4), 8(2, 1-3), 9(1), 10(1,. 1-2), 111), 141). Segment IV: hair 0(1), 1(1, 1-3),
2(1), 3(3, 1-4), 4(1, 1-2), 5(1), 6(1), 7(2, 1-3), 8(2, 1-2), 9(1), 10(2, 1-3), 11(1),
14(1). Segment V: hair 0(1), 1(1, 1-3), 2(1), 31, 1-2), 4(4, 2-5), 51), 6(1), 7(@3, 1-
4), 8(2, 1-3), 9(1), 10(1), 11(1), 14€1, 1-2). Segment VI: hair 0(1), 11, 1-2), 2(1),
3(1, 1-2), 4(2, 1-3), 5(1), 6(1), 701), 8@, 1-3), 9(1), 10€1), 11(1), 14(1). Segment
VII: hair 0(1), 1(€1), 2(1), 3(1), 4(€1), 5(1), 6(2, 1-4), 71, 1-2), 8@2, 1-2), 9(3, 2-4),
10(1), 11€1), 14(1). Segment VIII: hair 0(1), 4(1), 9(5, 4-6), 14(1, 1-2). Paddle: Uni-
formly lightly pigmented, midrib slightly darker; moderately elongate and oval,
sometimes slightly produced; external margin smooth; hair 1-P slender. Male genital
lobe extending to 0.47 of paddle, female genital lobe to 0.38.

LARVA (fig.28). Head: 0.88 mm. Siphon: 0.86 mm. Anal Saddle: 0.34 mm. Chae-
totaxy as given below, hairs moderately to darkly pigmented; in life, larvae are dark.
Diagnostic characters as in the key; readily separated from cozumelensis by the api-
cally rounded comb scales and hair 1-X single. Head: Width about 1.1 of length; in-
tegument strongly pigmented. Mental plate with strong median tooth and 10(9-11)
distinct teeth on either side, often a smaller detached tooth at base. Hair 0(1), 1(1),
3(1), 413, 12-16), 5(1), 6(1), 7(2, 2-3), 8(1, 1-2), 9(1, 1-2), 10(1), 11(6, 4-7), 12Q,
2-3), 13(1, 1-2), 14(3, 3-7), 15(2, 2-3). Antenna: Length about 0.33 of head; uni-
formly strongly pigmented. Hair 1-A about 0.54 from base. Thorax: Long hairs with
attenuate apices; long and stellate hairs weakly barbed. Prothorax: hair 0(7, 5-9),
1(3, 2-3), 2(1), 3(4, 3-5), 4(3, 3-5), 5(2, 2-3), 6(1), 7(3, 2-3), 8(6, 5-7), 9(1), 10(1),
11q1, 1-2), 12(1), 13(4, 3-5), 14(20, 18-24). Mesothorax: hair 1(5, 4-6), 2(1), 3(1),
4(1), 5(1), 6(4, 4-5), 7(1), 8(5, 4-6), 9(6, 4-7), 10(1), 11@, 1-2), 12(1), 13(6, 5-9),
14(5, 3-5). Metathorax: hair 1(4, 4-6), 2(1), 3(2, 1-3), 4(6, 3-6), 5(7, 6-8), 6(1), 7(6,
5-7), 8(5, 4-7), 9(5, 3-6), 10(1), 11(1, 1-2), 12(1), 13(6, 5-7). Abdomen: Long and
stellate hairs as on thorax. Segment I: 1(4, 4-6), 2(6, 5-7), 3(1), 4(3, 1-4), 5(7, 5-10),
6(3, 2-5), 7(1), 9(4, 3-5), 10(1), 11(6, 6-8), 13(3, 3-5). Segment II: hair 0(1), 1(4, 3-
6), 2(6, 6-7), 3(1), 4(2, 2-4), 5(7, 6-9), 6(3), 7(4, 3-4), 8(1, 1-2), 9(6, 4-7), 101),
11(1), 12(1), 13(5, 3-5). Segment III: hair 0(1), 1(3, 3-5), 2(7, 6-8), 3(1), 4(1), 5(7,
5-10), 6(2), 7(4, 3-5), 8(1), 9(7, 5-7), 10(1), 11(€1), 121), 133, 3-6), 14(1). Segment
IV: hair 0(1), 1(4, 1-4), 2(6, 6-7), 3(1), 4(1), 5(8, 6-9), 6(2), 7(4, 3-5), 8(1), 9(6, 5-
8), 10(1), 11(1), 121), 13(4, 3-4), 14(1). Segment V: hair 0(1), 1(2, 1-4), 2(6, 6-7),
3(1), 4(1, 1-2), 5(7, 6-8), 6(2), 7(4, 3-5), 8(1, 1-2), 9(4, 4-7), 10€1), 11€1), 121),
13(3, 2-4), 14(1). Segment VI: hair 0(1), 1(4, 2-5), 2(6, 5-7), 3(1), 4(1), 5(6, 5-8).
6(2, 1-2), 7(3, 2-3), 8(2, 1-2), 9(5, 4-6), 10(1), 11(1), 12(€1), 13(3, 3-4), 141, 1-2),
Segment VII: hair 0(1), 1(2, 2-4), 2(6, 5-7), 3(1), 4(1), 5(6, 5-7), 6(2, 2-4), 7(1), 8(4,
3-5), 9(3, 2-4), 10(1), 111), 12(1), 13(2, 1-3), 14(1, 1-3). Segment VIIT: Comb
scales in a single row, 16(14-19) in number; individual scale moderate, blunt apically
and finely fringed. Hair 0(1), 1(6, 5-7), 2(1), 3(7, 5-8), 41), 5(4, 3-5), 14(1, 1-2).
Siphon: Index about 2.3-2.7; darkly pigmented, imbricate. Pecten extending to 0.46;

56 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

teeth 14(12-17); smaller basally, progressively larger distad; individual tooth with a
long spine and a basal denticle. Hair 1(3, 2-3), 2(1), 3-5 not visible, 6(1), 7(1), 8(1),
9(1). Anal Segment: Saddle darkly pigmented, imbricate. Gills short, broad, unequal,
ventral 0.67 of dorsal. Hair 1(1, 1-3), 2(5, 4-6), 31), 4a(4, 4-6), 4b(4, 3-4), 4c(4, 3-
4), 4d(4, 3-5), 4e(3, 3-4), 4f(3, 1-3).

SYSTEMATICS. A. ioliota is extremely similar to cozumelensis from which it is
differentiated apparently only by the few constant diagnostic characters mentioned
above. There is no indication of any differentiation between the populations from
Trinidad and Venezuela. This species is apparently restricted to low and moderate
elevations in the Cordillera de la Costa in Venezuela and in Trinidad but may be
present in the intermediate area.

BIONOMICS. The reported breeding sites of ioliota are quite varied, ranging from
rockholes, treeholes, broken and cut bamboo to fallen cacao fruits and artificial con-
_tainers of varied size. Females have been taken in biting-landing collections (TR
1132). Attempts at experimental infection with yellow fever virus have yielded nega-
tive results (Aitken 1960:4).

DISTRIBUTION (fig.3). Trinidad and venesuel (northern) at low to moderate
elevations. Material examined: 115 specimens; 15 6, 23 ?, 30 pupae, 47 larvae; 30 in-
dividual rearings (12 pupal, 15 larval, 3 incomplete).

TRINIDAD. St. Andrew: Cumaca, elev. 250 ft, 22 Oct 1964, rockhole, A. Guerra (TR 783),
1 Ipd (783-104), 2 lp? (783-103,105); elev. 500 ft, 14 Jan 1965, tub, A. Guerra (TR 940), 3 Ip
(940-120,121,129), 3 Ip? (940-110,113,125), 1 p? (940-111), 3 IP (940-114,122,123), 12 L; 18
Feb 1965, tin can, A. Guerra, 1 Ip? (1008-21), 1 pd (1008-20), bamboo pot, 2 pd (1013-10,11),
1 p? (1013-13), 5 L (TR 1013) [UCLA]; Cumaca Road, elev. 500 ft, 14 Jan 1965, fallen cacao
pod, A. Guerra (TR 938), 4 lp? (938-128-131), 8 L [UCLA] . St. George: Mt. Beck, elev. 1000 ft,
29 Apr 1965, biting collection, A. Guerra, 1 @ (TR 1132) [UCLA] County not specified: F W.
Urich, 3 2, type series [USNM no. 16000].

VENEZUELA. Aragua: Choroni, Mar 1939, P.J. Anduze, 1 6 [USNM]. Guamito, 19 June 1927,
M.N. Tovar, 1 9; 4 July 1927, M.N. Tovar, 1 2 [USNM]; 1 Dec 1967, cut bamboo, R. Hansell and
J.J. Vera (VZ 83), 1 Ipd (83-13), 3 pd (83-101,105,106), 1 p? (83-102), 4 L [UCLA]. Maracay,
4 July 1927, M.N. Tovar, 1 pd [USNM]. Ocumare de la Costa, 12 Aug 1927, M.N. Tovar, 1 ?
[USNM]. Colonia Tovar, Capital, Guamitas, Rancho Grande and Uraca (Cova Garcia, Sutil and
Rausseo 1966b:73). Carabobo: Las Trincheras (Cova Garcia, Sutil and Rausseo 1966b:73). Fal-
con: Cumarebo (Cova Garcia, Sutil and Rausseo 1966b:73). Miranda: Carenero and Charallave (Co-
va Garcia, Sutil and Rausseo 1966b:74). Hacienda Los Chaguaramos (Suarez and Cova Garcia 1961:
17; Cova Garcia, Sutil and Rausseo 1966b:74). Los Teques and Petare (Cova Garcia, Sutil and
Rausseo 1966b:74). Trujillo: Sabana de Mendoza (Cova Garcia, Sutil and Rausseo 1966b:74).

13. Aedes (Howardina) cozumelensis Diaz Najera
Figs.3,8,29 30

1966.Aedes (Howardina) cozumelensis Diaz Najera, 1966a:331-344. TYPE: Holotype 6 with geni-
talia slide, San Miguel de Cozumel, Territorio de Quintana Roo, Mexico, 25 Sept 1965, A.
Diaz Najera [ISET, 6529].

Aedes (Howardina) cozumelensis of Stone (1967:208).

FEMALE. Wing: 3.5 mm. Proboscis: 2.98 mm. Forefemur: 2.28 mm. Abdomen:
about 2.5 mm. Very similar to ioliota in color and size, differing apparently only in
the following (fig.8). Hindtarsal segment 5 with some apical and posterior dark scales,
rest white.

Berlin: Subgenus Howardina 37

MALE. Wing: 3.15 mm. Proboscis: 2.45 mm. Forefemur: 2.01 mm. Essentially
similar to female in coloration.

MALE GENITALIA (fig.29). As figured; diagnostic characters as in the key. Ninth
tergite lobe indistinct, with 4,5 strong setae. Sidepiece length about 3.0 median
width. Claspette with 1 apical seta and varied number of smaller subapical setae, pro-
gressively smaller basally. Clasper about 0.5 of sidepiece, with 1,2 subapical hairs;
basal 0.5 with spicules.

PUPA (fig.29). Abdomen: 3.78 mm. Trumpet: 0.55 mm; index about 4.1. Paddle:
0.74 mm. Description based on a single skin. Diagnostic characters as in the key. Ap-
parently differing from ioliota by hair 1-II with a few simple branches and paddle
shorter, length about 1.2 greatest width. Cephalothorax: Mesonotum moderately
pigmented between trumpets, lateral areas and metanotum strongly pigmented; trum-
pets strongly contrasting. Hair 5-C strong, at least 2.0 as long as 4-C. Abdomen: Ter-
gite I strongly pigmented except for usual membranous areas; segments II-IV strong-
ly pigmented; rest of segments progressively lighter caudad. All large hairs except 1-
II, 9-VII,VIII usually single or double; hair 1-I] not distinctly dendritic and with
about 8 branches. Paddle: Uniformly moderately pigmented, midrib slightly darker;
slightly broadened in distal half; external margin with submarginal serrations. Male
genital lobe extending to about 0.5 of paddle, female genital lobe to 0.36.

LARVA (fig.30). Head: 0.76 mm. Siphon: 0.65 mm. Anal Saddle: 0.27 mm. De-
scription based on a single specimen. Diagnostic characters as in the key. Hairs mod-
erately to strongly pigmented. Readily separated from ioliota by the apically pointed
comb scales and hair 1-X double. Head: Hairs and branching as in ioliota. Mental
plate with a strong median tooth and 10 distinct teeth on either side, a smaller de-
tached tooth at base. Antenna: Slender, moderate; strongly pigmented. Hair 1-A at
about 0.56 from base. Thorax: Hair 13-P double. Segment VIII: Comb scales in a sin-
gle row, 14 in number; individual scale moderate, pointed apically and fringed. Si-
phon: Index about 2.0; imbricate, moderately pigmented, with darker basal ring.
Pecten extending to 0.6; teeth 15,16; individual tooth with long spine and a small
basal denticle. Anal Segment: Saddle imbricate, moderately pigmented and with
darker basal ring. Gills short, broad, unequal, ventral about 0.64 of dorsal.

SYSTEMATICS. A. cozumelensis is obviously very closely related to ioliota from
which it is separated in all stages apparently only by the few constant features men-
tioned in the diagnosis. This species is known from the Yucatan Peninsula and the
adjacent Cozumel Island and British Honduras.

BIONOMICS. The immature stages of cozumelensis have been found only in arti-
ficial containers but it is very likely that the natural breeding sites of this species are
similar to those of the closely related ioliota. Adults of cozumelensis have been col-

‘lected at dusk on human bait in British Honduras.

DISTRIBUTION (fig.3). Yucatan Peninsula and adjacent islands in Mexico (states
of Quintana Roo and Yucatan) and British Honduras. Material examined: 11 speci-
mens; 1 6, 8 9, 1 pupa, | larva; no individual rearings.

BRITISH HONDURAS. Cayo: Georgeville, elev. 300 ft, 27 July 1967, D.S. Bertram, biting col-
lections, 1 ? (BH A369); 1 Aug 1967, D.S. Bertram, 3 ? (BH A339) [UCLA]. Roaring Creek, elev.
70 ft, 27 July 1967, biting collections, D.S. Bertram, 1 + (BH A339); 3 Aug 1967, DS. Bertram,
1 2 (BH 376); 9 Aug 1967, D'S. Bertram, 1 9 (BH 427) [UCLA].

MEXICO. Quintana Roo: Cozumel Island, 25 Sept 1965, A. Diaz Najera, holotype 6 (6529),
I paratype 2, 1 p. Yucatan: Tizimin, 8 Dec 1960, M. Munoz, 1 L [ISET].

58 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Albonotatus Group

FEMALES. Head: Narrow decumbent scales white. Palpus with a few apical sil-
very white scales. Thorax: Inner dorsocentral line narrow, creamy; median prescutel-
lar line broad, white; outer dorsocentral line narrow, pale golden; lateral marginal line
narrow, silvery white, extending from side of anterior promontory to wingbase.
Acrostichal bristles not developed on disc. Scutellum predominantly dark scaled,
light scales restricted to median lobe. Ssp with distinct silvery patch; upper stp patch
separate, small. Legs: Foretarsus with a few basal white scales on segment 2; midtar-
sus with small incomplete basal white rings on segments 1,2; hindtarsus with broader
complete basal white rings on segments 1-3. Abdomen: Basomedian tergal light
markings creamy on segments II-VII or III-VII.

MALES. Palpus with basomedian white scales on segments 4,5. Larger claw of
foreleg and midleg with small acute basal tooth and large blunt submedian tooth;
smaller claw simple.

MALE GENITALIA. Claspette represented te a basomesal tubercle bearing | api-
cal bristle. Clasper subequal to or definitely shorter than di cartons subapical hairs
not developed. Aedeagus ovate, subequal to proctiger.

PUPAE. Cephalothorax: Hair 5-C double to 4-branched, subequal to hair 4-C and
distinctly shorter than distance from its base to that of trumpet; hair 7-C simple to
triple; hair 10-C single to 5-branched. Abdomen: All large hairs except 1-II, III, 9-VII,
VIII usually single or double; hair 2-II laterad of or in line with hair 3; hair 5-IV,V
short, not exceeding apex of tergite following; hair 9-II-VI varied in position; 9-VII
longer, stouter and with more branches than 6-VII. Paddle: Very long or moderate;
apex rounded or slightly emarginate; marginal spicules distinct.

LARVAE. Head: Hair 1-C unbranched; 4-C moderately long, multiple; 5-C single,
long; 6-C usually double, strong; none of the ventral hairs stellate; 14-C with a few
spikelike branches; 15-C double, long, exceeding apex of mental plate. Antenna:
Shaft without distinct spicules; hair 1-A very thin, single. Thorax: Hair 13-P absent.
Only hairs 8,13-P distinctly stellate; hair 1-T far mesad of hairs 2,3. Abdomen: Integ-
ument glabrous; hair 2 distinctly displaced cephalad but not beyond hair 8, hair 9
not markedly displaced cephalad. Only hairs 2,9-I-VI, 11-I distinctly stellate, with a
-few spikelike branches; hair 8-VII distinctly cephalad of 9-VII. Segment VIII: Comb
scales in a single or partial double row; individual scales rather short, blunt and faint-
ly fringed. Siphon: Index less than 2.5; integument without spicules; distinct free or
attached acus present. Pecten not extending beyond middle of siphon; teeth numer-
ous, closely set and with basal denticles. Anal Segment: Saddle body without distinct
spicules; caudal margin with very short pointed spines preceded by a few spicules;
hair 1-X usually double (1-3), near lower end of caudal margin. Ventral brush with 6
pairs of hairs on a moderately or strongly sclerotized boss with distinct grid bars, 1
or 2 hairs usually detached.

DISCUSSION. The 2 very similar species of the Albonotatus Group are strikingly
differentiated from each other by male genitalic characters, an unusual occurrence in
the subgenus. The group is very similar in adult ornamentation to the Ioliota Group,
differing from it primarily in having hindtarsal segment 5 completely dark. The lar-
vae of the Albonotatus Group lack hair 13-P which is present in the Ioliota Group.
Since the loss of hair 13-P is a derived feature it appears that the Albonotatus Group
was probably derived from the loliota Group.

The Albonotatus Group has a distribution complementary to that of the Busckii
Group (fig.3). A. albonotatus has been reported from Hispaniola (Haiti and Domini-

Berlin: Subgenus Howardina 59

can Republic) and doubtfully from Guadeloupe, and bahamensis from the Bahamas
(Grand Bahama and New Providence). As indicated under albonotatus, a third spe-
cies of the group may be present in the Dominican Republic.

The natural breeding sites of the group are probably treeholes and broken bam-
boo, as has been reported for albonotatus. To date bahamensis has been found only
in artificial containers.

14. Aedes (Howardina) albonotatus (Coquillett)
i Figs.3,8,31

1905.Gymnometopa albonotata Coquillett, 1905:183. TYPE: Lectotype 6 (141.3) with associated
larval and pupal skins, San Francisco Mts., Santo Domingo [Dominican Republic] , 3 Sept
1905, A. Busck [USNM, 8297; selection of Stone and Knight, 1956:213].

Aedes (Howardina) albonotatus of Dyar (1928:233-234); Edwards (1932:155); Gerry (1932:40);
Lane (1939:98; 1953:717-719); Horsfall (1955:471); Stone and Knight (1956:213); Stone,
Knight and Starcke (1959:173); Belkin, Schick and Heinemann (1965:18); Forattini (1965:
461); Porter (1967:38).

Aedes (Howardina) albonotata of Bonne and Bonne-Wepster (1925:369-370).

Aedes albonotata of Dyar and Knab (1906b:190,196); Dyar (1906:15); Howard, Dyar and Knab
(1912:70; 1917:618,619,834-835); Dyar (1918:73,80); Root (1927:465).

Gymnometopa albonotata of Coquillett (1906b:25); Theobald (1907:211; 1910:219); Surcouf
and Gonzalez-Rincones (1911:212,213).

Haemagogus albonotata of Dyar and Knab (1906a: 166).

FEMALE. Wing: 2.9 mm. Proboscis: 2.1 mm. Forefemur: 1.91 mm. Abdomen:
about 2.98 mm. As described for the group, with following apparently diagnostic
features. Head: Median stripe on vertex narrow, white; light scales on sides and ven-
ter straw yellow. Erect scales predominantly dark, a few median ones whitish. Palpus
about 0.14 of proboscis. Thorax (fig.8): Scales on anterior promontory white. Inner
dorsocentral line narrow, creamy, reaching prescutellar space, followed by a white-
scaled median prescutellar line; outer dorsocentral line narrow, golden; lateral mar-
ginal line narrow, pure white, extending from sides of anterior promontory to wing-
base, apparently interrupted only at scutal angle. Antealar area above paratergite
with scattered dark scales. Light scales on median scutellar lobe narrow, white; lateral
lobes with a few narrow bronzy scales. Pleural scales as in the section and group.
Legs: Forefemur and midfemur predominantly dark, a few basal scales creamy, prox-
imal 0.5 of posterior surface mostly straw yellow, extending distally as narrow line,
rest dark; hindfemur straw yellow to proximal 0.5 on both sides, rest dark scaled.
Hindtarsal segment 3 white scaled in at least basal 0.7. Haltere: Stem pale, knob
mostly whitish, a few dark scales dorsally. Abdomen: Basomedian tergal light scales
dingy white on II-VII. Sternites straw yellow to segment IV; rest with basal straw
yellow and apical dark scales.

MALE. Wing: 2.28 mm. Proboscis: 1.93 mm. Forefemur: 1.49 mm. Essentially
similar to female in coloration.

MALE GENITALIA (fig.31). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with 4 strong bristles. Sidepiece length about 3.0 median
width; most tergomesal bristles parallel to longitudinal axis. Clasper narrow, long,
subequal to sidepiece; spiniform about 0.2 of clasper. Dorsal sclerotization of aedea-
gus slightly emarginate.

60 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

PUPA (fig.31). Abdomen: 2.7 mm. Trumpet: 0.35 mm; index 4.5. Paddle: 0.7 mm.
Chaetotaxy as given below. Diagnostic characters as in the key. Readily separated
from bahamensis by longer paddle, hair 7-C single and 9-VIII only 3-branched. Ceph-
alothorax: Mesonotum and metanotum lightly pigmented; trumpets contrasting. Hair
L(1).22(2)s. Shlout- 2) AO 2-3): 5222-3). 601d: 44) BOQ), 9C1), FOC), LIC ye l2ds:
1-2). Abdomen: Tergite I lightly pigmented except for usual membranous areas; pig-
mentation of other segments progressively lighter caudad. Segment I: hair 1(9-13 pri-
mary branches, each with many secondary ones), 2(1), 3(1), 4(6, 2-8), 5(2, 1-2),
6(1), 7(1), 9€1). Segment II: 0(1), 1(2, 2-3), 21), 31), 4@2, 1-3), 5C1, 1-2), 6(1),
7(1, 1-2), 9(1). Segment III: hair 0(1), 1(1, 1-2), 21), 31), 4@, 2-3), 5(1), 6(1),
7(2, 1-3), 8(1, 1-2), 9(1), 10(1), 11(1), 14(1). Segment IV: hair 0(1), 1@, 1-2), 2(1),
3(1, 1-2), 4(2), 5(2), 6(1), 7(2, 1-2), 8(1), 9(1), 10(1), 11€1), 141). Segment V: hair
01), 11; 1-2), 21), 30); 401, 1-2), 51), 6G): 7@,y 1-2),-8Q; 1-2), 9G); 10),
11(1), 14(1). Segment VI: hair 0(1), 1(1), 2(1), 3(1), 41), 5(1), 6(1), 7(1), 8@, I-
2), 9(1), 10(1), 11(€1), 14(1). Segment VII: hair 0(1), 1€1), 21), 301), 4C1), 5(),
6(1), 71), 8(1), 9(2, 2-3), 10(1), 111), 14(1). Segment VIII: hair 0(1), 4(1), 9(3),
14(1). Paddle: Long; uniformly lightly pigmented, midrib slightly darker; external
margin serrated; hair 1-P slender. Male genital lobe extending to 0.55 of paddle, fe-
male genital lobe to 0.34.

LARVA (Description based on Howard, Dyar and Knab, 1917:853; only frag-
ments of two individual rearings seen). Diagnostic characters as in the key. Readily
separated from bahamensis by the subequal gills and nature of the comb scales. Head:
Width slightly greater than length. Mental plate broadly triangular, with a median
tooth and 12 distinct teeth on either side. Antenna: Slender, moderate, uniform; all
hairs single. Thorax: Rounded, wider than long. Abdomen: Moderate, anterior seg-
ments shorter. Segment VIII: Comb scales in a single row of about 15; individual
scale elongate, rounded at tip and with small spinules. Siphon: Short, tapered on
outer half, with an index of about 2.0. Pecten extending beyond middle, hair 1-S just
within the last tooth; individual tooth with long spine and a small basal denticle.
Anal Segment: Saddle as long as wide; caudal margin fringed with spines. Gills about
2.0 length of anal segment; broad and saclike, ventral pair subequal to dorsal.

SYSTEMATICS. In spite of the very limited material available and the very close
similarity in the ornamentation of. the adults, I consider albonotatus to be undoubt-
edly a species distinct from bahamensis. The male genitalia of the 2 species are mark-
edly differentiated and the immature stages are also obviously distinct.

A. albonotatus has been reported from Haiti, Dominican Republic, Bahamas and
Guadeloupe (Stone, Knight and Starcke 1959:173). I have not seen specimens from
Haiti but this record may be correct as the type locality of albonotatus is in the Do-
minican Republic, also on the island of Hispaniola. The record from the Bahamas is
based on a single female from New Providence which in my opinion is bahamensis
(see). The male from Guadeloupe collected by Busck [USNM] is unquestionably al-
bonotatus and it is possible that this species does occur on this island. However, no
additional specimens of albonotatus have been found in extensive surveys on Guade-
loupe in recent years and the possibility exists, particularly in view of the extreme
similarity of this specimen with males from the type series of albonotatus from the
Dominican Republic, that an error occurred in labeling Busck’s material and that this
Guadeloupe specimen was actually part of this type series. In addition to the type se-
ries I have seen 2 females from the Dominican Republic, near Sanchez [USNM],
which differ from the former in the presence of some pale scales on the remigium of
the wing. These females are only tentatively identified as albonotatus since they are

Berlin: Subgenus Howardina 61

from the northern part of the complex island of Hispaniola and there is a possibility
that another species of the group may occur there.

The record of albonotatus from Ecuador (Levi-Castillo 1952a:77; 1952b:555;
1953:40) is probably based on a misidentification of an undescribed species of the
Ioliota Group (see).

BIONOMICS. The type series of albonotatus was reared from larvae and pupae
collected in bamboo stumps.

DISTRIBUTION (fig.3). Island of Hispaniola (Haiti and Dominican Republic) and
doubtfully Guadeloupe. Material examined: 17 specimens; 5 6, 6 9, 4 pupae, 2 larvae;
4 individual rearings (2 pupal, 2 larval).

DOMINICAN REPUBLIC. Sanchez, 5 Apr 1949, J.M. Brennan, 2 9? [USNM]. San Francisco
Mts, Sept 1905, bamboo, A. Busck, 1 Ipd (141-2), 1 lp? (136-2), 1 pd (136-3), 1 p? (141-7), 1¢
(141), 1 6, 1 9 (136), 1 9 [USNM, UCLA].

?GUADELOUPE. July 1905, A. Busck, 1 6 [USNM].

HAITI. Mt. Cabrite, 12 Oct 1924, W.A. Hoffman (Root 1927:465).

15. Aedes (Howardina) bahamensis Berlin, n.sp.
Figs.3,8 32,33

TYPES: Holotype 6 (BAH 1) with genitalia slide, village of Hunter, Grand Bahama, Bahama Is-
lands, 28 Nov 1964, A. Spielman and A.E. Weyer, from colony established from larvae collected in
artificial containers [USNM]. Allotype 2 (BAH 1), same data as holotype. Paratypes: 10 6,5 2,9
pupae and 12 larvae (BAH 1), same data as holotype [UCLA].

Aedes (Howardina) albonotatus of Spielman and Weyer (1965:339-343); in part of Stone, Knight
and Starcke (1959:173).

FEMALE. Wing: 3.5 mm. Proboscis: 2.62 mm. Forefemur: 2.36 mm. Abdomen:
about 3.25 mm. Essentially similar to albonotatus but differing by a slightly larger
size and in the following (fig.8). Palpus about 0.16 of proboscis. Hindtarsal segment
3 white scaled in barely more than basal 0.5. Basomedian tergal light scales on abdo-
men creamy on II-VII. Sternites completely straw yellow up to V, rest with basal
straw yellow and apical dark scales.

MALE. Wing: 3.06 mm. Proboscis: 2.45 mm. Forefemur: 2.28 mm. Essentially
similar to female in coloration.

MALE GENITALIA (fig.32). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with 4 strong setae. Sidepiece length about 3.0 median width;
basolateral scales white, remainder dark; tergomesal surface clothed with numerous
strong, long setae; distal sternomesal margin with threadlike setae with twisted apices.
Clasper broad at base, curved mesad on distal 0.7. Dorsal sclerotization of aedeagus
deeply emarginate, producing distinct apical lobes.

PUPA (fig.32). Abdomen: 4.05 mm. Trumpet: 0.54 mm; index about 4.0. Paddle:
0.78 mm. Diagnostic characters as in the key. Chaetotaxy as given below, hairs dark-
ly pigmented. Readily distinguished from albonotatus by the shorter paddle, 7-C usu-
ally double and 9-VIII at least 5-branched. Cephalothorax: Mesonotum moderately
pigmented between trumpets; sides and metanotum deeper; trumpets strongly con-
trasting. Hair 1(1, 1-2), 2(2, 2-4), 3(1), 4(4, 3-6), 5(2, 2-4), 6(1, 1-2), 7(3, 1-3), 8(3,
2-5), 9(1, 1-2), 10(4, 2-5), 11(1), 12(3, 2-5). Abdomen: Tergite I uniformly strongly
pigmented except for usual membranous areas; other segments imbricate, strongly
pigmented, progressively lighter caudad. Segment I: hair 1(3-10 primary branches,

62 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

each with many secondary ones), 2(1), 3(1), 4(5, 4-6), 5(5, 3-6), 6(2, 1-3), 7(3, 2-4),
9(1), 10(1, present in many), 11(1, present in a few). Segment II: hair 0(1), 1(6, 4-
14), 2(1), 31), 4(6, 4-8), 5(1, 1-2), 6(@, 1-3), 7(3, 2-5), 9(1), 10(2, 1-2, present in a
few). Segment III: hair 0(1, 1-2), 1(2, 1-4), 2C), 3(1, 1-2), 4(3, 3-5), 5(1, 1-2), 6(3,
1-4), 7(4, 2-9), 8(3, 2-4), 9(1), 10(2, 2-3), 11€1), 141). Segment IV: hair 0(1), 1€1,
1-2), 2(1), 3(3, 2-4), 4Q, 1-3), 51, 1-2), 6@, 1-3), 7(3, 2-4), 8(3, 2-5), 9(1), 10@,
1-3), 11(1), 12(1, present in one specimen), 14(1). Segment V: hair 0(1, 1-2), 1(2, 1-
3), 21), 3(1, 1-3), 4(3, 2-5), 5(1, 1-2), 6(2, 1-2), 7(3, 2-4), 8(3, 2-4), 9(1), 10(1, 1-
2), 11€1), 14(1). Segment VI: hair O(1, 1-2), 11, 1-3), 2(1), 3(2, 1-2), 4(2, 2-3),
5(1), 6(1), 7(2, 1-3), 8(3, 2-4), 9(1), 101, 1-2), 1101), 141, 1-2). Segment VII: hair
Odi, 1-2), 101, 1-3), 2(1), 3@, 1-3), 4€1, 1-2), 51), 6(1, 1-2), 701), 8(2, 1-3), 9(@3, 2-
5), 101), 1101), 14€1). Segment VIII: hair 0(1), 4(1), 909, 5-12), 14(1, 1-2). Paddle:
Uniformly lightly pigmented, midrib slightly darker; dorsal surface with fine den-
ticles; external margin distinctly serrated and faintly emarginate; hair 1-P slender.
Male genital lobe extending to 0.65 of paddle, female genital lobe to 0.44.

LARVA (fig.33). Head: 0.92 mm. Siphon: 0.73 mm. Anal Saddle: 0.35 mm. Chae-
totaxy as given below, hairs darkly pigmented. Diagnostic characters as given in the
key. Readily separated from albonotatus by the unequal gills, pecten teeth termi-
nating before middle of siphon and hair 1-S distinctly beyond last pecten tooth.
Head: Width about 1.06 of length, moderately pigmented, deeper around mouth-
parts. Mental plate with a strong median tooth and usually 12(11-13) distinct teeth
on either side, often a smaller detached tooth at base. Hair 0(1), 1(1), 3(1), 4Q9, 7-
10), 51), 6(2, 1-2), 7(2, 2-3), 8(1), 9(2, 1-3), 10(1), 1105, 3-6), 12(2, 2-3), 131),
14(2, 2-3), 15(2, 2-3). Antenna: Length about 0.34 of head; moderately pigmented
and without spicules; hair 1-A at about 0.5 from base. Thorax: Apices of long hairs
attenuate; stellate hairs with a few branches; long and stellate hairs weakly barbed.
Prothorax: hair 0(6, 4-6), 1(3, 3-4), 2(1), 3(4, 3-5), 4(3, 2-5), 5(2, 2-3), 6(1), 7(2, 2-
3), 8(7, 5-8), 9(2, 1-2), 10(1), 11(2, 1-3), 12(1), 14(20, 17-29). Mesothorax: hair
1(3, 3-4), 2(1), 3(1), 4€1), 5C1, 1-2), 6(5, 4-5), 7(1), 8(6, 5-9), 9(5, 5-8), 10(1), 111,
1-2), 12(1), 13(6, 3-7), 14(3, 3-4). Metathorax: hair 1(4, 3-5), 2(1), 3(3, 3-5), 43,
2-4), 5(3, 3-5), 6(1), 7(6, 5-7), 8(6, 3-6), 9(5, 4-6), 10(1), 111, 1-2), 121), 13(4, 4+
7). Abdomen: Stellate hairs as on thorax. Segment I: hair 1(6, 4-7), 2(7, 6-8), 3(1),
4(6, 4-8), 5(6, 4-9), 6(4, 3-5), 7(1), 9(4, 3-5), 10(1), 11(5, 4-7), 13(4, 2-4). Segment
II: hair 0(1), 1(3, 2-4), 2(8, 6-8), 3(1), 4(6, 4-6), 5(5, 3-6), 6(4, 3-5), 7(2, 1-3), 8(2),
9(4, 3-5), 10(1), 11€1, 1-2), 12(1), 13(4, 3-5). Segment III: hair 0(1), 1(3, 2-3), 2(7,
6-8), 3(1), 4(2, 1-3), 5(4, 3-7), 6(3, 2-3), 7(5, 4-7), 8(1), 9(6, 5-7), 101, 1-2), 11d],
1-2), 121, 1-2), 13(3, 2-4), 14(1). Segment IV: hair 0(1), 1(2, 2-3), 2(7, 5-8), 31),
4(2, 1-3), 5(5, 4-7), 6(3, 2-4), 7(5, 4-6), 8(1, 1-2), 9(6, 6-9), 10(1, 1-2), 11(1, 1-2),
12(1), 13(2, 2-4), 14(1). Segment V: hair 0(1), 1(3, 2-3), 2(8, 6-9), 3(1), 4(@, 2-3),
5(6, 3-9), 6(2, 2-3), 7(5, 4-5), 8(1, 1-2), 9(7, 6-8), 10(1), 11(2, 1-2), 12(1), 13@, 2-
3), 141, 1-2). Segment VI: hair 0(1, 1-2), 1(3, 2-5), 2(6, 6-8), 3(1), 4(1), 5(4, 3-5),
6(2, 1-2), 7(4, 3-6), 8(2, 2-3), 9(5, 5-7), 10€1), 11(2, 1-3), 12(1), 134, 2-5), 14(@, 2-
3). Segment VII: hair 0(1), 1(4, 3-6), 2(7, 6-8), 3(1), 4C1), 5(4, 3-4), 6(3, 2-4), 71,
1-2), 8(4, 4-5), 9(3, 3-4), 10€1, 1-2), 11€1, 1-2), 12(1), 13(3, 1-4), 14(3, 2-3). Seg-
ment VII: Comb scales in a single row, in the form of a wide chevron, 10-18 in num-
ber; individual scale moderate; apex blunt and finely fringed. Hair 0(1), 1(4, 2-5),
2(1), 3(7, 6-8), 4(1), 5(4, 2-4), 142, 2-3). Siphon: Index about 2.18-2.4; integument
imbricate, moderately pigmented, with darker basal ring. Pecten extending to 0.4,
usually 10(6-14) teeth, smaller at base, progressively larger distad; individual tooth
with long spine and a basal denticle; hair 1-S distinctly distad of last tooth; acus dis-

Berlin: Subgenus Howardina 63

tinct, free or attached. Hair 1(3, 2-4), 2(1), 3-5 not visible, 6(1), 7(1), 8(2, 2-3), 9(1).
Anal Segment: Saddle moderately pigmented and imbricate; caudal margin with
short spines. Gills short to moderately long, unequal, ventral about 0.5-0.6 of dorsal.
Hair 1(2, 1-3), 2(1), 3(4, 4-6), 4a(4, 3-5), 4b(3), 4c(3, 2-4), 4d(3, 3-4), 4e(3, 3-4),
4f(3, 2-4).

EGG. Length about 0.67 mm. Width about 0.14 mm; exochorion transparent,
with characteristic circular clear refractive bodies; endochorion dark and with large
irregular polygonal sculpturing.

SYSTEMATICS. I have no hesitation in recognizing bahamensis as a species dis-
tinct from the superficially similar albonotatus since the male genitalia are complete-
ly different in the 2 forms and the immature stages also appear to be differentiated
by constant features mentioned in the diagnosis. In addition to the type series de-
rived from a laboratory colony established from material collected on Grand Baha-
ma Island and described as albonotatus by Spielman and Weyer (1965:339-343), I
have included under bahamensis a single female reared from a treehole on New Prov-
idence Island, on which previous records of albonotatus from Bahamas were based.
Although this specimen agrees well with the type material, the presence of bahamen-
sis on New Providence can be confirmed only by additional material, especially males
and immature stages.

BIONOMICS. Spielman and Weyer (1965) recorded bahamensis (as albonotatus)
from a wide variety of artificial containers similar to those utilized by Aedes aegypti
but never in association with the latter. The natural breeding sites of bahamensis are
probably treeholes, as indicated for the female from New Providence Island, and pos-
sibly also rockholes. According to Spielman and Weyer (loc. cit.) females of baha-
mensis are normally autogenous but will occasionally take blood from man or canar-
ies. A laboratory colony was readily established and maintained; eggs produced by
autogenous females hatched in tap water after storage for 1-2 weeks at a relative hu-
midity of 76% and a temperature of 76°F.

DISTRIBUTION (fig.3). Bahama Islands. Material examined: 45 specimens; 13 d,
8 2,9 pupae, 15 larvae; no individual rearings.

BAHAMA ISLANDS. Grand Bahama: village of Hunter, artificial containers, A. Spielman (BAH-
1, type series), 11 6,6 2,9 p, 12 L [USNM, UCLA]; 19 July 1964, tire, A. Spielman, 1 6 genitalia
(GB 71-1), 1 L; 1 Dec 1964, tire. A. Spielman, 1 6 (GB 81-52A), 1 9 (GB 81-85A) [USNM] . New
Providence: 1915, treehole, H.G. Dyar, 1 9 [USNM].

Busckii Group

FEMALE. Head: Narrow decumbent scales pale golden. Palpus completely dark
scaled. Thorax: Inner dorsocentral line faint, narrow, golden; median prescutellar
line narrow, golden; outer dorsocentral line faint, narrow, golden; lateral marginal
line narrow, silvery, interrupted at scutal angle and sometimes caudad of it. Acros-
tichal bristles present on disc. Scutellum with some light scales on lateral lobes. Ssp
with distinct silvery patch; upper stp patch small, separate. Legs: Foretarsus com-
pletely dark; midtarsus with a few basal white scales on segments 1,2; hindtarsus
with short to moderate basal white rings on segments 1-3. Abdomen: Basomedian
tergal light markings evident from segment V distad.

MALE. Palpus entirely dark. Larger claw of foreleg and midleg with submedian
tooth only; smaller claw without either submedian or basal tooth.

MALE GENITALIA. Claspette with a distinct outstanding apical fingerlike cylin-

64 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969 —

drical process more than twice as long as wide and bearing a short apical bristle.
Clasper about 0.6 of sidepiece; subapical hairs absent. Aedeagus ovate; apex deeply -
emarginate, with distinct lateral lobes; subequal in length to proctiger.

PUPA. Cephalothorax: Hair 5-C double to 4-branched, short but longer than hair
4-C; hair 7-C double to 4-branched; hair 10-C with 2-5 branches. Abdomen: All large
hairs except 1-II, III, 9-VII, VIII usually single or double; hair 2-II laterad of or in line
with hair 3; hair 5-IV,V short, usually not reaching apex of tergite following; hair 9-
II-VI usually caudad of level of hair 6; 9-VII longer, stouter and with more branches
than hair 6-VII; hair 4-VIII usually single. Paddle: Short; apex rounded; marginal
spicules distinct.

LARVA. Head: 1-C unbranched; 4-C moderately long, multiple; 5-C single, long;
6-C usually double or triple; hairs 11,14-C weakly stellate, 11-C with unequal branches;
15-C usually double, long, exceeding apex of mental plate. Antenna: Shaft with a
few scattered weak sharp spicules; hair 1-A single. Thorax: Hair 13-P absent. Hairs
0,8,14-P, 1,13,14-M, 1,4,5,8,13-T all moderately stellate and moderately pigmented;
hairs 1-M,T far mesad of hairs 2,3. Abdomen: Integument glabrous; hairs 2 and 9 far
removed cephalad but not beyond hair 8. Hairs 1,2,5,9,13-I-VI, 11-1, 7-III-VI all
moderately stellate, usually stronger than on thorax; hair 8-VII cephalodorsad of 9-
VII. Segment VIII: Comb scales varied from 1 more-or-less regular row to 2-4 dis-
tinct rows; individual scales moderately long and broad, blunt apically and fringed.
Siphon: Index less than 3.0; integument without spicules; small separate acus pres-
ent. Pecten variable, sometimes extending to near apex; teeth always few and widely
spaced, short, proximal with basal denticles. Anal Segment: Saddle body without
distinct spicules; caudal margin with relatively short fringed spines; hair 1-X usually
less than 5-branches, near lower end of caudal margin. Ventral brush with 5 pairs of
hairs on a moderately sclerotized boss with short median grid bars.

DISCUSSION. Only 1 species is recognized at this time in the Busckii Group al-
though it is evident, as indicated under the species, that there are marked differences
among the populations from different islands.

The Busckii Group is immediately separated from all the other groups in the sub-
genus by the presence of acrostichal bristles on the disc of the mesonotum. The or-
namentation of the adults is in general intermediate between the Walkeri Group and
the Ioliota-Albonotatus Groups. The larvae of some populations show another prim-
itive feature in the presence of several rows of comb scales, a condition found nor-
mally only in the Fulvithorax Group. Since the distributions of these 2 groups are
now adjacent and may have overlapped in the past, the comb patch in busckii may
have been acquired through introgression from the Fulvithorax Group. This does not
seem likely because this feature is found only in the northern populations, farthest
removed from Fulvithorax Section and because there is no indication of any other
introgression between the 2 groups.

The Busckii Group is definitely known only from the Lesser Antilles from Sint
Eustatius southward to Grenada (fig.3). This distribution is complementary to the
Ioliota and Albonotatus groups except for a possible overlap with the latter in Gua-
deloupe.

The Busckii Group has the widest range of breeding sites in the subgenus, being
found in treeholes, broken bamboo, rockholes, fallen fruits and leaves, as well as in
leaf axils of bromeliads and aroids, and in flower bracts of heliconias.

Berlin: Subgenus Howardina 65

16. Aedes (Howardina) busckii (Coquillett)
Figs.3,8,34,35

1906.Stegomyia busckii Coquillett, 1906:60-61. TYPE: Lectotype d, Cacao plantation, San Do-
mingo [Dominica], 28 July 1905, A. Busck [USNM, 9139; selection of Stone and Knight
$956: 215).

1919.Aedes thaxteri Dyar and Knab, 1919:8-9. TYPE: Holotype 6, Grand Etang, Grenada, Nov
1912, R. Thaxter [USNM, 21704] . Synonymy with busckii by Dyar and Knab (1919:8).

1941.Aedes tracei Senevet and Quievreux, 1941:251-258. TYPE: Holotype 6 (M 2163) with asso-
ciated pupal skin and genitalia mount, near Piscine Colson, Martinique, 14 Aug 1939, G.
Senevet and L. Quievreux [NE; see Belkin 1968]. Synonymy with busckii by Belkin, Schick
and Heinemann (1965:33).

Aedes (Howardina) busckii of Bonne and Bonne-Wepster (1925:372-373); Dyar (1928:234-235);
Edwards (1932:155); Lane (1939:99); Floch and Abonnenc (1945:14-16,42,45,47); Lane
(1953:722-724); Stone, Knight and Starcke (1959:173); Belkin, Schick and Heinemann (1965:
17,33); Porter (1967:38).

Aedes busckii of Dyar and Knab (1906b:189,192); Howard, Dyar and Knab (1917:860-862); Dyar
(1918:73,80); Senevet (1938:189); van der Kuyp (1954:39,64,68,70,72,95,96); Stone and
Knight (1956:215).

Gymnometopa busckii of Coquillett (1906b:25); Theobald (1907:211; 1910:219); Surcouf and

_ Gonzalez-Rincones (1911:212,213). |

Haemagogus busckii of Coquillett (1905 :183); Dyar and Knab (1906a: 166).

Aedes (Howardina) thaxteri of Bonne and Bonne-Wepster (1925:276-277); Lane (1939:99).

Aedes (Soperia) tracei of Lane (1953:703-705); Stone, Knight and Starcke (1959:173); Forattini
(1965 :460). |

FEMALE. Wing: 3.24 mm. Proboscis: 2.19 mm. Forefemur: 2.19 mm. Abdomen:
about 3.0 mm. As described for the group and with following additional features.
Head: Light scales on median stripe pale golden; creamy on sides and venter. Thorax
(fig.8): Light patch on anterior promontory silvery. All mesonotal light lines, except
lateral marginal, narrow, golden. Antealar area above paratergite with scattered dark
scales. Lateral scutellar lobes with a few narrow golden and dark scales. Legs: Fore-
femur and midfemur predominantly dark anteriorly, a few creamy scales at base,
posterior surface with a lateral and a median longitudinal creamy line from base to
apex on forefemur and midfemur respectively, rest dark; hindfemur predominantly
creamy to 0.6, lateral margin and distal area dark. Haltere: Stem pale, knob predom-
inantly dark, fringe scales whitish. Abdomen: Sternites completely creamy on II-IV;
rest of segments with basal creamy and apical dark scales.

MALE. Wing: 2.54 mm. Proboscis: 2.19 mm. Forefemur: 1.84 mm. Essentially
similar to female in color.

MALE GENITALIA (fig.34). As figured; diagnostic features as described for group.
Ninth tergite lobe indistinct, with 2,3 strong bristles. Sidepiece elongate, length about
3.5 median width.

PUPA (fig.34). Abdomen: 3.35 mm. Trumpet: 0.43 mm; index about 4.0. Paddle:
0.7 mm. Chaetotaxy as given below, hairs moderately pigmented. Diagnostic charac-
ters as in the key. Cephalothorax: Mesonotum lightly to moderately pigmented be-
tween trumpets; lateral areas and metanotum deeper; trumpets strongly contrasting.
Hair 1(1), 21, 1-2), 3(1), 403, 2-4), 5@, 2-4), 6(1, 1-2), 7@2, 2-4), 8@, 1-3), 9(1),
10(3, 2-5), 11(1), 12(2, 1-3). Abdomen: Tergite I moderately pigmented and imbri-
cate except for usual membranous areas; pigmentation progressively lighter on distal

66 | Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

segments. Segment I: hair 1(9-14 primary branches, each with many secondary ones),
2(1, 1-2), 3(1), 4(4, 3-5), 5(3, 2-6), 6(1, 1-2), 7@, 1-3), 91), 111, present in a few).
Segment II: hair 0(1), 1(4, 3-7), 21, 1-2), 31, 1-2), 4(4, 2-5), 5C1, 1-3), 61, 1-2),
7(2, 1-3), 8(1, 1-3, present in many), 9(1), 11(1, present in some). Segment III: hair
O(1), 1(2, 1-3), 21, 1-3), 31, 1-2), 4(3, 2-4), 51, 1-4), 6(1, 1-2), 7(3, 2-4), 8(2, 2-3),
9(1), 10(2, 1-2), 11(1), 14(1). Segment IV: hair 0(1), 1(2, 1-3), 2(1), 3(3, 2-4), 4Q,
2-3), 5(2, 1-3), 6(1), 702, 2-4), 8(2, 1-3), 9(1), 10(2, 1-3), 11(€1), 14(1). Segment V:
hair O(1), 11, 1-2), 2(1), 3@, 1-3), 4(3, 3-5), 5(1, 1-3), 6(1), 7(3, 2-3), 8(2, 1-4),
9(1), 10(1), 11(1), 14(1). Segment VI: hair 0(1), 1(1, 1-2), 2(1), 3(@2, 1-3), 4(@2, 2-3),
S(1, 1-2), 6(1), 7(2, 1-3), 8(2, 1-3), 9(1), 10€1), 11(1), 14(1). Segment VII: hair 0(1),
Id, 1-2), 2(1), 3(1), 4@, 1-2), 5(1, 1-2), 6(1, 1-3), 7€1), 8(1, 1-2), 9(5, 3-7), 10C1,
1-2), 11(1), 14(1). Segment VIII: hair 0(1), 4(1, 1-2), 9(7, 5-11), 141, 1-2). Paddle:
Uniformly lightly pigmented, midrib slightly darker; length subequal to or slightly
greater than median width; dorsal surface with minute denticles; apex rounded, ex-
ternal margin serrated; hair 1-P slender. Male genital lobe extending to about 0.45 of
paddle, female genital lobe to 0.36. :

LARVA (fig.35). Head: 0.73 mm. Siphon: 0.59 mm. Anal Saddle: 0.27 mm. Diag-
nostic characters as in the key. Readily separated from other members of the section
by the presence of only 5 pairs of hairs in ventral brush. Chaetotaxy as given below,
hairs moderately to darkly pigmented. Head: Width about 1.08 of length; integument
moderately pigmented, deeper around mouthparts. Mental plate with a strong medi-
an tooth and usually 12(10-13) distinct teeth on either side, often a smaller detached
tooth at base. Hair 0(1), 1(1), 3(1), 4€11, 10-12), 5(1), 6(2, 2-3), 7(3, 2-5), 8(1), 91,
1-3), 10(1), 11(5, 4-10), 12(2, 2-3), 13(1), 14(3, 2-6), 15(2, 2-4). Antenna: Length
about 0.3 of head; uniformly moderately pigmented, with a few scattered spicules.
Hair 1-A at about 0.6 from base. Thorax: Apices of long ha?rs attenuate; stellate hairs
with a few to a moderate number of branches; long and stellate hairs weakly barbed.
Prothorax: hair 0(7, 6-9), 1(4, 2-5), 2(1), 3(5, 4-8), 4(4, 4-5), 5(2, 2-3), 6(1),.7(2),
8(7, 5-12), 9(2, 1-3), 10(1), 11(3, 2-3), 12(1), 14(20, 19-29). Mesothorax: hair 1(8,
4-12), 2(1), 3(1), 4(1), 51), 665, 3-5), 7(1), 865, 5-7), 9(6, 4-8), 10(1), 11(1, 1-3),
12(1), 13¢5, 4-8), 14(7, 3-10). Metathorax: hair 1(6, 3-12), 2(1), 3(2, 1-3), 4(7, 4-
10), 5(7, 5-10), 6(1), 7(5, 5-7), 8(5, 4-6), 9(4, 4-6), 10(1), 11(2, 1-3), 12(1), 1366,
6-13). Abdomen: Integument and stellate hairs as on thorax. Segment I: hair 1(7, 4-
11), 2(10, 7-13), 3(1), 4(2, 1-3), 5(7, 7-12), 6(5, 3-6), 7(1), 9(4, 3-5), 10(1), 1109,
7-12), 13(5, 3-6). Segment II: hair 0(1), 1(8, 5-12), 2(12, 7-12), 3(1), 4(2, 1-2), 5(9,
7-11), 6(4, 3-6), 7(4, 2-4), 8(1), 9(5, 4-7), 10(1), 11€1), 12(1), 13(7, 4-12). Segment
Hit: hair 0(1), 1(7, 5-12), 210, 8-10), 3(1), 4(1), 5(8, 8-14), 6(2, 2-3), 7(5, 4-7), 8(1),
9(9, 6-10), 10(1), 11€1), 12(1), 13(8, 4-12), 14(1, 1-2). Segment IV: hair 0(1), 1(10,
7-10), 2(10, 7-10), 3(1), 4(1), 59, 8-13), 6(2), 7(5, 4-7), 8(1, 1-2), 9(8, 8-10), 10(1),
11(1), 12(1), 13(8, 5-12), 14(1). Segment V: hair 0(1), 1(5, 5-12), 2(10, 7-10), 3(1),
4(1, 1-2), 5(8, 8-12), 6(2), 7(5, 4-8), 8(1), 9(9, 7-10), 10(1), 11(1), 12(1), 13(7, 5-
11), 14€1, 1-2). Segment VI: hair 0(1), 1(9, 5-12), 2(8, 5-10), 3(1), 4(1), 5(9, 7-13),
6(2), 7(3, 3-5), 8(2, 1-3), 9(7, 6-10), 10(1), 11(1), 12(1), 13(5, 3-10), 14(2, 1-2).
Segment VII: hair 0(1), 109, 4-12), 2(7, 5-10), 3(1), 4(1), 5(7, 5-11), 63, 3-5), 7(1,
1-2), 8(6, 5-8), 9(3, 2-4), 10(1), 11(1), 12(1), 13(3, 3-7), 14(1). Segment VIIT: Comb
scales varied from a single regular row to a distinct patch of 3,4 rows; individual
scale long, blunt apically and fringed. Hair 0(1), 1(10, 6-12), 2(1), 3(7, 6-7), 4(1),
5(6, 3-10), 14(1, 1-2). Siphon: Index about 2.2-2.8; integument moderately pigmen-
ted, with darker basal ring. Pecten teeth few (6-10), widely spaced; frequently teeth
terminate before hair 1-S, occasionally 1,2 teeth extend beyond it; individual tooth

Berlin: Subgenus Howardina 67

with a long spine and small basal denticle; acus distinct and separate or indistinct.
Hair 1(2, 2-4), 2(1), 3-5 not visible, 6(1), 7(1), 8(1), 9(1). Anal Segment: Saddle
moderately pigmented; caudal margin with moderately large flat spines; individual
spine with lateral spinules. Gills short to moderate, broad, unequal, ventral about 0.6
of dorsal. Hair 1(2, 2-4), 2(4, 4-6), 3(1), 4a(4, 3-6), 4b(3, 3-4), 4c(4, 3-4), 4d(4, 4-5),
4e(2, 2-3). |

SYSTEMATICS. A. busckii in the present interpretation is undoubtedly a complex
of closely related and more or less strongly differentiated forms, some of which may
be geographically or ecologically isolated species. It is impossible to resolve this com-
plex at this time because of the bewildering array of differentiation and variation in
ornamentation of the adults and larval characters which is apparently associated with
both isolation on numerous islands and the utilization of a wide variety of breeding
sites (see Bionomics below). As in the case of the 7ripteroides caledonicus complex
in the South Pacific (Belkin 1955), the situation may be further complicated through
recent mixing of originally isolated insular populations by human transport.

For the present, the currently accepted synonymy of thaxteri with busckii is re-
tained although there are rather obvious differences in ornamentation of the adults
as pointed out in the original description of the former. The synonymy of tracei,
proposed by Belkin, Schick and Heinemann (1965:33), appears to be well founded
since the populations from Martinique are very similar to those from Dominica.

The above diagnosis of the complex is based primarily on topotypic associated ma-
terial of all stages of busckii from Dominica. In the adults, the most striking differ-
ences in the various populations are: (1) in the nature of the lateral mesonotal silvery
line which may be nearly continuous from the anterior promontory to the wing root
or variously broken into segments, (2) in the extent of the basal white ring on hind-
tarsal segment 3 which varies from less than 0.4 to almost 0.75 of the segment length,
and (3) the presence of some white scales on additional tarsal segments. No striking
differences have been noted in the pupae but the larvae of different populations show
remarkable differentiation from the typical population in: (1) the number and ar-
rangement of scales in the comb from a few scales in a single even row to a large
number in a patch of 3 or 4 rows, and (2) the distribution of the pecten which may
end before the siphonal tuft or may reach nearly to the tip of the siphon.

The northern populations from Sint Eustatius, St. Kitts, Montserrat and Guade-
loupe differ from all the others in the more extensive white ring on hindtarsal seg-
ment 3, the larval comb in a large patch and the pecten ending before the siphonal
hair tuft. The male genitalia of these populations are also apparently distinct from
the others in the large number of setae on the tergal surface of the sidepiece and in a
broad aedeagus with distinct apicolateral lobes. Although these populations probably
represent a distinct species, I prefer not to name or describe it formally without addi-
tional material and a thorough study of the entire complex which is beyond the
scope of the present revision.

The busckii complex is apparently restricted to the main chain of the Lesser An-
tilles from Sint Eustatius southward to Grenada. I have seen material only from the
above-mentioned islands and from St. Kitts, Montserrat, Guadeloupe, Dominica,
Martinique and St. Lucia. No material of the complex was found in a recent survey
of Nevis and St. Vincent but it is likely that populations will be found there as well
as on the smaller intermediate islands of the chain. The report of busckii from Jamai-
ca by Hill and Hill (1948:43) is probably based on an erroneous identification of a
member of the Aurites Group and its record from the Dominican Republic is due to
the error in labeling the type material San Domingo instead of Dominica (see How-

68 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

ard, Dyar and Knab 1917:860).

BIONOMICS. Members of the busckii complex have been found breeding in the
entire range of habitats utilized by the subgenus, in rockholes, treeholes, broken and
cut bamboo, fallen leaves, fruits and pods, in the leaf axils of bromeliads and aroids,
and in the flower bracts of heliconias. The immature stagés are also commonly found
in a variety of artificial containers. One male (FWI 245) from Guadeloupe is re-
ported from a ground pool but this may be a contamination or an error in labeling.
Females are frequently taken in biting-landing collections and some were obtained
on chicken bait. Apparently some adult populations may reach high levels and be-
come serious pests locally.

DISTRIBUTION (fig.3). Main chain of the Lesser Antilles from Sint Eustatius to
Grenada; reported from low (40 ft) to moderate (2600 ft) elevations. Material exam-
ined: 1269 specimens; 253 6, 273 9, 290 pupae, 453 larvae; 175 individual rearings
(19 pupal, 131 larval, 25 incomplete).

DOMINICA. St. Andrew: Calibishie, 23 June 1965, colocasia leaf axil, R. Martinez and A. Guer-
ra, 1 p, 21 (DOM 137), 2 lpd (138-10,11), 1 lp? (138-13), 1 P (DOM 138) [UCLA]. St. David:
Castle Bruce, 19 Apr 1959, bamboo, R.H. Darsie, 1 Ipd (76-14), 2 pd (76-15A,16), 1 d (76-10),
1 lp? (76-6), 4 2 (76-2,4A,7,11), 1 IP, 1 P [USNM]. Grand Fond, elev. 300 ft, 18 June 1965, biting
collection, R. Martinez and A. Guerra, 1 9? (DOM 116) [UCLA]. Terre Ferme, elev. 1800 ft, 22
June 1965, heliconia bract, R. Martinez and A. Guerra (DOM 131), 1 lpd (131-12), 1 Ip? (131-11),
6 6,12 p,2L [UCLA]. St. George: L’Etang, elev. 2600 ft, 8 June 1965, bromeliads, T.H.G. Ait-
ken, R. Martinez and A. Guerra, 1 p (DOM 53) [UCLA]. Trafalgar Falls, elev. 1700 ft, 9 June
1965, biting collection, R. Martinez and A. Guerra, 4 ? (DOM 58); heliconia bract (DOM 59), 2 lp?
(59-10,11), 1 P, 1 p [UCLA]. St. Joseph: Batati Estate, elev. 40 ft, 14 June 1965, treehole, R. Mar-
tinez and A. Guerra (DOM 82), 1 Ip? (82-20) [UCLA]. Clarke Hall Estate, 12 Nov 1964, P.J. Span-
gler, 1 9; 11 Mar 1964, D.F. Bray, 2 9; 21-31 Jan 1965, W.W. Wirth, 3 ?; 24 Jan 1965, W.W. Wirth,
2 2;3 May 1966, R.J. Gagne, 1 9; 11 May 1966, G. Steyskal, 1 9; 19 May 1966, G. Steyskal, 1 9;
24 May 1966, G. Steyskal, 3 2; 1 June 1966, G. Steyskal, 1 9; 3 June 1966, G. Steyskal, 1 2; 10
June 1966, G. Steyskal, 6 9; 11 June 1966, G. Steyskal, 2 9; 12 June 1966, G. Steyskal, 1 9; 10
Oct 1966, A.B. Gurney, 1 9 [USNM]. St. John: Milton Estate, elev. 1000 ft, 24 June 1965, heli-
conia bract, A. Guerra (DOM 159), 2 Ips (159-10,13), 1 pd (159-101), 2 P, 4 p, 14 L [UCLA].
St. Luke: Point Michele, Champigny Estate, elev. 600-700 ft, 6 June 1965, treehole, T.H.G. Aitken,
R. Martinez and A. Guerra, 2 Ipd (15-20,22), 1 Ip? (15-23), 1 P, 1 p, 1 1(DOM 15), 1 lpd (19-10),
4 6,8 p, 7 L(DOM 19), 4 L (DOM 20), fallen leaf, 1 Ip? (17-10), 2 p, 6 L, 21(DOM 17), steel
drum, 3 6, 1 2? (DOM 21), heliconia bract, 4d, 1 9? (DOM 23) [UCLA]. St. Mark: Tete Morne, elev.
1400 ft, 10 June 1965, heliconia bract, R. Martinez and A. Guerra (DOM 68), 2 Ipd (68-12,14),
2 Ip? (68-11,13), 4 6, 8 p, 4 L [UCLA]. St. Paul: Antrim Valley, 7 June 1965, treehole, R. Mar-
tinez and A. Guerra, 4 9, 4 p, 4 L (DOM 35) [UCLA]. Mount Joy, elev. 1000 ft, 7 June 1965,
treehole, R. Martinez and A. Guerra (DOM 41), 1 Ipd (41-15), 2 Ip? (41-10,11), 1 P, 11 L [UCLA].
Pont Casse, 23 Nov 1964, P.J. Spangler, 1 6 [USNM]. Parish not specified: July 1905, in sugar
boiler full of vegetation, 1 6, 1 2 (78), type series [USNM no. 9139].

GRENADA. St. Andrew: Balthazar Estate, elev. 450 ft, 15 Oct 1963, bamboo, R. Martinez,
1 Ipd (43-125), 1 Ipd (44-136), 1 Ip? (44-131) [UCLA]. Mt. St. Catherine, elev. 2000 ft, 25 Oct
1963, bamboo, R. Martinez, 2 L (GR 82), heliconia bract, 2 ? (GR 85) [UCLA]. St. David: Minor-
ca Estate, elev. 700 ft, 24 Oct 1963, colocasia leaf axil, R. Martinez, 3 L (GR 76); 1 Nov 1963, fal-
len cacao pod, R. Martinez (GR 111), 1 lpd (111-18) [UCLA]. St. George: Annandale, elev. 700 ft,
25 Aug 1929, leaf axil, F.M. Root, 18 6, 24 9? (LAR 40), 5 2? (LAR 40A); 28 Oct 1963, bamboo,
R. Martinez, 1 L (GR 95), rockhole (GR 96), 1 Ip (96-156) [UCLA]. Grand Etang, 2 July 1929,
F.M. Root, 20 6, 40 9 (GRR 11); 26 Aug 1929, leaf axil, F.M. Root, 21 6,4 ? (LAR 42); 17 Oct
1963, heliconia bract, R. Martinez (GR 54), 1 Ip? (54-132), bamboo (GR 59), 1 Ipd (59-133),
fallen palm leaf, 9 6,3 2 (GR 60), heliconia bract (GR 61), 1 Ipd (61-127), 1 Ip? (61-139); 18 Oct
1963, biting collection, R. Martinez, 1 9 (GR 62); 23 Oct 1963, heliconia bract, R. Martinez (GR
67), 1 lpdé (67-140), 1 6 [UCLA]. St. Mark: Concord, F.M. Root, 27 d, 15 2? (GRR 80) [UCLA].

Berlin: Subgenus Howardina 69

Parish not specified: FM. Root, 11 6, 29 ? (GRR 40) [UCLA].

GUADELOUPE. Basse-Terre: Capesterre, 76 L, Matouba, 72 L; Petit-Bourg, 17 L; Pointe-Noire,
12 L (Floch and Abonnenc 1945:16). Route de Traversee, km 4-6, elev. 200 ft, 18 Sept 1965,
ground pool, P. Fauran, 1 6 (FWI-245) [UCLA]. St. Claude, Bains Jaunes, elev. 900 ft, 31 Mar
1965, heliconia bract, P. Fauran, 2 Ipd (211-12,13), 5 lp? (211-10,11,15-17), 2 IP (211-14,18),
4 Ipd (212-11,13,15,16), 2 Ip? (212-10,14), 1 IP (212-12) [UCLA]. Soufriere, 111 L; Trois-
Rivieres, 51 L (Floch and Abonnenc 1945:16). Vernou, elev. 300 ft, 20 July 1964, treehole, P.
Fauran (FWI 182), 4 lpd (182-101,102,105,107), 6 lp? (182-103,106,108,110,112,113), 1 IP (182-
111), 2 L [UCLA]. Locality not specified, elev. 3000 ft, July 1905, wild canna, A. Busck, 1 Ipd
(80-10), 2 ? (80-3,7), 36,49 [USNM].

MARTINIQUE. Fort de France, 1929, bamboo, F.M. Root, 2 6 (LAR 23), bromeliads, 1 9
(LAR 28A) [UCLA]; Route de la Trace, elev. 680 ft, 16 July 1965, heliconia bract, P. Fauran
(FWI-M2), 2 Ipd (M2-10,11), 1 Ip? (M2-14), 3 pd (M2-15,101,102), 2 p? (M2-12,100), 1 IP (M2-
13),3 p, 11 [UCLA]. Mt. Pelee, July 1905, A. Busck, 2 9? [USNM]. Piscine Colson, 14 Aug 1939
(Senevet and Quievreux 1941:263).

MONTSERRAT. St. Antony: Chance Mountain, elev. 2600 ft, 5 Oct 1966, treehole, T.H.G.
Aitken, R. Martinez and A. Guerra, 2 P, 10 L(MNT 25), 2 lpd (28-10,11), 1 lp? (28-13), 1 pd (28-
100), 2 IP (28-12,14), 19 L (MNT 28), 1 Ipd (29-10), 2 lp? (29-11,12), 3 L (MNT 29), heliconia
bract (MNT 27), 4 Ipd (27-12,13,15,16), 2 lp? (27-10,14), 1 pd (27-100), 1 IP (27-17), 5 L[UCLA].
Parsons Estate, elev. 500 ft, 13 Oct 1966, tin can, R. Martinez and A. Guerra (MNT 72), 1 IP (72-
10),9 L [UCLA]. Soufriere, elev. 1000 ft, 12 Oct 1966, bamboo, R. Martinez and A. Guerra, 1 IP
(66-10), 1 p, 5 L(MNT 66), 5 L (MNT 67) [UCLA]; Lower Gages, 1250 ft, 6 Oct 1966, treehole,
R. Martinez and A. Guerra, 8 L (MNT 36) [UCLA] ; Upper Gages, 1750 ft, 6 Oct 1966, heliconia
bract, T.H.G. Aitken, R. Martinez and A. Guerra (MNT 38), 1 Ipdé (38-10), 1 pd (38-100), 1 p?
(38-101), 5 L [UCLA]. Waterworks Estate, elev. 550 ft, 18 Oct 1966, treehole, R. Martinez and
A. Guerra (MNT 118), 1 p? (118-100), 1 IP (118-20) [UCLA]. St. Peters: Plymouth, Belham
Bridge, elev. 250 ft, 4 Oct 1966, tin can, T.H.G. Aitken, R. Martinez and A. Guerra (MNT 20),
2 Ipd (20-11,12), 1 Ip? (20-10), 1 pd (20-100), 1 p? (20-101), treehole (MNT 22), 1 lpd (22-14),
3 Ip? (22-10-12), 1 IP (22-13), 9 6, 3 9, 12 p [UCLA].

SINT EUSTATIUS. 7 Oct 1947, E. van der Kuyp, 2 6, 1 9 [USNM].

ST. KITTS. Christ Church: Old Road Town, elev. 1200 ft, 5 Sept 1966, heliconia bract, A.
Guerra, 1 6, 2 9; 5 Sept 1966, R. Martinez and A. Guerra, 1 p? (93-101), 2 6, 1 9 (KIT 93), 1 6,
4? (KIT 94) [UCLA]. St. John: Dieppe Bay Town, slopes of Mount Misery, elev. 2500 ft, 17 Aug
1966, treehole, T.H.G. Aitken, R. Martinez and A. Guerra (KIT 25), 1 lp? (25-10), 1 IP (25-11),
Oc, 2 %,3 pi. (UCLA):

ST. LUCIA. Anse la Raye: Elev. 300 ft, 15 July 1964, treehole, R. Martinez and A. Guerra,
1 L (LU 30); 30 July 1964, fallen fruit, A. Guerra, 1 lp? (125-20), 1 p? (125-100), 1 lp (126-20),
4 L(LU 126), treehole (LU 128), 7 Ip? (128-10-16), 1 IP (128-17), 26 6, 10 9, 19 P, 11 L [UCLA];
Derriere Dos, 31 July 1964, fallen cacao pod, A. Guerra (LU 131), 1 lpd (131-24), 5 1p? (131-20-
23,25), 4 6, 12 ?, 8 p [UCLA]. Barre de L’Isle, elev. 800-900 ft, 14 July 1964, treehole, T.H.G.
Aitken, R. Martinez and A. Guerra, 3 lpd (14-14-16), 6 lp? (14-10-13,18,19), 8 p, 7 L, 3 1(LU 14),
2? (LU 16), 1 ? (LU 18); 1 Aug 1964, treehole, R. Martinez and A. Guerra, 3 L (LU 143); 6 Aug
1964, bamboo, R. Martinez and A. Guerra, 14 L(LU 159), 20 L(LU 160), 3 lpd (161-12,13,14),
1 IP (161-13), 3 p (LU 161) [UCLA]. Castries: Bise Quarry, elev. 250 ft, 17 July 1964, treehole,
R. Martinez and A. Guerra (LU 46), 3 lp? (46-20,22,23), 2 IP (46-21,24), 3 L, 11 [UCLA]. Cul de
Sac Estate, 30 July 1964, chicken bait, R. Martinez and A. Guerra, 2°? (LU 129) [UCLA]. Delgos,
elev. 750 ft, 14 July 1964, bamboo, T.H.G. Aitken, R. Martinez and A. Guerra, 3 lpd (10-11,13,
15), 2 Ip? (10-12,14), 1 IP (10-16), 1 P, 7 L(LU 10), 3 lp? (11-13-15), 1 p? (11-101), 3 IP (11-10,
12,18), 1 p, 7 L (LU 11), 3 Ip? (12-10-12), 3 L (LU 12), treehole (LU 13), 2 L, 1 1 [UCLA].
Grande Anse, elev. 175 ft, 20 July 1964, heliconia bract, R. Martinez and A. Guerra, 1 L (LU 62),
bamboo (LU 63), 1 Ipdé (63-11), 4 lp? (63-10,12-14), 1 pd (63-101), 2 L, 1 1 [UCLA]. La Croix
Junction near Poix Bouche, elev. 500 ft, 20 July 1964, treehole, R. Martinez and A. Guerra (LU
59), 1 Ipd (59-10), 6 L [UCLA]. Marquis Estate, elev. 200 ft, 20 July 1964, fallen cacao pod, R.
Martinez and A. Guerra, 1 6, 1 ?, 1 p, 3 L (LU 60), fallen cocoanut (LU 61), 2 lp? (61-10,13),

70 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

1 p? (61-11), 16 L [UCLA]. Piton Flor, elev. 1875 ft, 4 Aug 1964, bamboo, R. Martinez and A.
Guerra (LU 146), 1 lpé (146-10), 1 L [UCLA]. Union Agricultural Station, elev. 200 ft, 16 July
1964, bamboo, R. Martinez and A. Guerra (LU 42), 1 Ip? (42-10), 3 IP (42-11,13,14), 10 L
[UCLA]. Waterworks Road, 20 July 1964, bromeliads, R. Martinez and A. Guerra (LU 67), 1 p?
(67-100) [UCLA]. Micoud: Mahaut, elev. 700 ft, 27 July 1964, heliconia bract, A. Guerra (LU
101), 2 Ipd (101-14,15), 2 Ip? (101-12,13), 1 IP (101-11), 9 L [UCLA] . Mon Repos, elev. 300 ft,
28 July 1964, bromeliads, A. Guerra (LU 104), 1 Ipd (104-11), 6 L [UCLA] . Soufriere: Soufriere,
elev. 220-500 ft, 20 July 1964, heliconia bract, R. Martinez and A. Guerra, 2 L (LU 78), bamboo
(LU 80), 4 lpé (80-10-13), 2 Ip? (80-14,15), 10 L; 22 July 1964, fallen cacao pod, R. Martinez and
A. Guerra (LU 83), 1 lp? (83-11), 3 L [UCLA]. 3

SEXLINEATUS SECTION

FEMALES. Medium sized to rather large species, wing 2.7-5.0 mm. Head: Subme-
dian and lateral dark patches broad, distinct. Erect scales restricted to occiput. Torus
dark brown to almost black, with light scales restricted usually to mesal surface. Pal-
pus with or without apical light scales. Thorax: Mesonotum always with distinct lin-
ear light pattern, light lines of same color throughout, white, creamy or golden (ex-
cept in ecuadoriensis). Acrostichal line not developed but a small patch of light
scales present at anterior promontory; inner and outer dorsocentral lines usually
long, both reaching scutellum; lateral marginal line usually continuous (broken in
Whitmorei Group); median prescutellar line not developed. Paratergite with scales.
Acrostichal bristles absent on disc. Pleural scales silvery, white or creamy; ppn and
ssp with scales; psp usually with scales (usually absent in Allotecnon Group); mep
with separate upper and lower patches. Legs: Midfemur and hindfemur sometimes
with distal white spots on anterior surface. All tibiae usually completely dark (ex-
cept Sexlineatus and Whitmorei Groups). Tarsal light markings not restricted to seg-
ment 1. Wing: Light scales present on remigium in Aurivittatus Group, absent in
others. Abdomen: Sternites II-IV with dark scales distally.

MALES. Larger claw of foreleg and midleg with submedian tooth only.

MALE GENITALIA. Lobes of tergite IX either poorly defined or distinct, bearing
a few to many strong to moderate setae. Sidepieces separated from each other at
base by a distance less than diameter of aedeagus at that level (greater in Whitmorei
Group); tergomesal and/or sternomesal margins with lobes only in Whitmorei Group.
Claspette completely absent in Whitmorei Group, represented in other groups by
sternal basomesal lobe not connected across midline and bearing 1 apical seta.

PUPAE. Cephalothorax: Hair 5-C longer than distance between its base and that
of trumpet, strong and always double, at least 2.0 of hair 4-C. Abdomen: Hair 5-IV,
V strong, usually double, extending distinctly beyond apex of tergite following; 2-
IV-VI usually within basal 0.6 of its tergite; hair 9-III-VI distinctly cephalo-laterad of
hair 6; 9-VII moderate to strongly developed, removed considerable distance cepha-
lad of hair 6.

LARVAE. Stellate hairs always strongly developed. Antenna: Short to moderate;
shaft with scattered spicules; hair 1-A usually 2,3-branched. Thorax: Integument
with or without spicules. Tubercles of hairs 5-7-P distinctly joined; hair 9-P with
more than 4 branches and faintly stellate; 13-P present, distinctly stellate. Abdomen:
Integumentary spiculation strong or weak, sometimes confined to dorsal or ventral
surface, spicules usually hairlike (spinelike in Allotecnon Group). Hairs 1,2,5,11,13-
I, 1,2,5,7,9 (and usually 13)-II-VII stellate; hair 8-VII cephalodorsad or caudodorsad

Berlin: Subgenus Howardina Tt

of 9-VII. Segment VIII: Hair 1,5-VIII usually stellate. Comb scales in a single row;
individual scales long, pointed apically. Siphon: Index 2.0-4.3; integument smooth,
imbricate or spiculate. Acus distinct, separate or attached. Pecten teeth with basal
denticles. Anal Segment: Saddle with distinct marginal spicules. Hair 1-X moderate,
with few branches; 3-X single or double; 2-X multiple. Ventral brush with 5 or 6
pairs of hairs arising from a boss with distinct grid bars. |

DISCUSSION. The ornamentation pattern of the adults of the Sexlineatus Sec-
tion is superficially so similar in most species that few specimens have been correctly
identified in the past and many distinct species were not recognized until in the pres-
ent study correlated differences were found in the immature stages. On the whole,
the section is a homogeneous one and strongly differentiated from the other 2 sec-
tions in all stages as indicated in the keys and diagnosis. As far as is known, members
of this section breed only in bromeliad axils.

The Sexlineatus Section is primarily associated with the great cordilleras of Cen-
tral America and western South America. Apparently only the monotypic Sexlinea-
tus Group and the Whitmorei Group penetrate eastward into the lowlands of the
Orinoco and the smaller northern South American drainages and no species extends
even to the Guianas. There are only small extensions from the Andes down into mod-
erate elevations near the headwaters of the Orinoco, Amazon and La Plata basins
(fig.1).

Only 2 of the 6 groups recognized here for the 19 species are strongly differenti-
ated: (1) the Quadrivittatus Group which is confined to Middle America and appears
to have retained the most significant primitive features in the section, and (2) the.
Whitmorei Group which is known in Colombia, Venezuela and Bolivia but undoubt-
edly also occurs in Ecuador and Peru; this is the most derived group in the section.
The remaining 4 groups show considerable overlap in many characters although the
Sexlineatus and Allotecnon Groups each have a characteristic facies in the adult and
larval stages. The Eleanorae Group in many respects appears intermediate between
the Quadrivittatus and the Allotecnon Groups. The separation of the Aurivittatus
Group from the Eleanorae Group is based only on the presence of light scales on the
remigium of the wing of the former; when the immature stages are known the 2 taxo-
nomic groups may have to be merged.

Two of the groups, Quadrivittatus and Allotecnon, are confined to Middle Ameri-
ca where they are sympatric and extend northward to southern and central Mexico.
The monotypic Sexlineatus Group overlaps with the above throughout their range in
Middle America but is apparently not completely sympatric with them; it further ex-
tends along the Pacific coast to the Gulf of Guayaquil and along the southern Carib-
bean shores to Trinidad; it is the only species of the section to be found near sea
level. The Whitmorei, Eleanorae and Aurivittatus Groups are confined to South
America. The dominant Eleanorae Group is primarily Andean and has a wide distri-
bution from western Venezuela and Colombia down to Bolivia, with slight intrusions
into the upper Orinoco and Amazon basins. The Aurivittatus Group is known only
from the upper drainages of the Amazon and La Plata in Bolivia and northern Argen-
tina respectively. The Whitmorei Group has been found in Colombia, Venezuela and
Bolivia and has small intrusions into the upper Orinoco, Amazon and La Plata drain-
ages.

Both in Middle and South American groups, some members of a given group seem
to occur sympatrically, others are allopatric and still others appear to be ecologically
isolated (in the Eleanorae Group) although broadly sympatric.

72

2(1).

3(2).

4(1).

5(4).

6(5).

7(A4).

8(7).

Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

KEYS TO GROUPS AND SPECIES

ADULTS

Midfemur without any indication of a white or silvery spot or streak on an-
terior surface above ventral light line (Quadrivittatus Group) . . . . .2
Midfemur with a more or less distinct white or silvery spot or streak on ante-
Wor SUriace gbove vet light lime . 8 Pe ee

Quadrivittatus Group

Outer dorsocentral line not reaching lateral marginal line anteriorly
19. brevis; 20. spinosus
Outer dorsocentral line ‘connected to lateral ‘marginal line anteriorly ts

Ppn scales predominantly narrow and yellowish; ssp scales in a restricted lin-
ear patch not connected to ppn patch; first flagellar segment with dark
pedics 9. 1 0, -, 3.027 quadrivittatus

Ppn scales predominantly broad and white: ssp scales extending toward ppn
patch, the 2 patches usually connected; first flagellar segment with white
Oe . 18. lorraineae

Antealar area (above paratergite) with numerous dark bronzy scales below

narrow supraalar golden line (Sexlineatus and Whitmorei Groups) . . . 5
Antealar area with whitish, yellowish or golden scales only, so that lateral
Marginal anc siuptaaiar lines are united .... 8 er a

Sexlineatus Group

Outer dorsocentral line reaching anterior border of mesonotum; torus with
small mesal patch of white or whitish scales. . . . . . .21. sexlineatus
Outer dorsocentral line not reaching anterior border of mesonotum, directed
outward toward lateral mesonotal line in fossa; torus with a broad ring of

white or silvery scales interrupted only dorsally (Whitmorei Group) . .6
Whitmorei Group
Hindtarsal white rings on segments 1,2 and 3; anterior end of outer dorsocen-
tral line with a patch of broad silvery scales. . . . . . .22. whitmorei
Hindtarsal white rings on segments | and 2 only; anterior end of outer dorso-
central line without patch of broad silvery scales. . 23. pseudodominicii
Base of vein R (remigium) with white or whitish scales . . ..... .8
Base of vein K with dark scales as elsewhere; .. 2 2 ee

Aurivittatus Group

Hindtarsal white ring on segment 1 only; haltere knob with white scales only
. .35. vanemdeni

Hindtarsal white eee on ‘sestients 1 2 and - haltere knob with both dark
and light seales. 3 5. ; ae Oo ee a Pee ee ke

Berlin: Subgenus Howardina 73

9(8). Outer dorsocentral line reaching anterior border of mesonotum; lateral mar-
ginal line poorly developed in front of scutal angle, without broadened

SCaleS 4 5: Coa SS. QUTIVETalGs

Outer dorsocentral line not teachin ariterior ‘border of mesonotum; lateral
marginal line strongly developed in front of scutal angle, with broadened

SCALES ui aoe Gr we Hinahs ls 5 ee en ne a

Eleanorae Group

10(8). Hindtarsus with basal white rings on segments 1 and 3 only. . 27. eleanorae
Hindtarsus with basal white rings on segments 1,2, 1-3 or 14... .. 11
11(10).Hindtarsus with basal white rings on segments eS wee Beet. 4 ie
Hindtarsus with basal white rings on segments 1,2 or 1-3 only. ie OLS

12(11).Inner dorsocentral line white, usually very narrow; ppn with narrow curved
light scales in upper part and broad flat white scales in lower; ssp scale
patch extended upward toward ppn broad scales; hindtarsal markings
broad, on segment 4 about 0.5 of segment length . . .31. ecuadoriensis
Inner dorsocentral line yellowish, usually broadened anteriorly; ppn predom-
inantly with moderately broad, curved, light scales throughout; ssp scale
patch entirely linear, not extended to ppn scales; hindtarsal markings nar-
rower, on segment 4 usually less than 0.2 of segment length
aes brevivittatus

13(11).Hindfemur without silvery spot on the anterior surface; ssp scale patch nar-
row, linear and not extended toward apn or ppn scale patches. . . . 14
Hindfemur with a very distinct silvery spot on dark portion of anterior sur-

face not connected to basal ventral light streak; ssp scale patch short and

broad and extended toward apn and ppn scale patches (Allotecnon Group)

16

14(13).Midfemoral light patch or spot at about0.5. .... . . . . 30. osornoi
Madfemoral light patch.at about 0.67 of distad 2.6.6. dc nei euetante . > FS
15(14).Outer dorsocentral line strongly broadened anteriorly and joining lateral me-
SOUOtal ite 2... ok tis wt ae, merinkellel

Outer dorsocentral line not strongly broadened. anteriorly and not joining

Valeral TICROMOta) ANG boca cr ee ak a a ae ee

Allotecnon Group

16(13).Hindtarsus with basal white rings on segments 1 and 2 only.
20. guatemala
Hindtarsus with basal white rings on n seements is. tg ae ee

17(16).Ppn with pure white broad flat scales except for narrow upper border of yel-
lowish curved moderately broad scales; psp scales usually present.

co 2, Swerrero

Ppn predominantly with yellowish, narrow to moderately broad, curved

74

2(1).

3(1).

4(3).

5(4).

6(5).

7(6).

8(7).

(5).

10(9).

Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

scales throughout except at lower posterior angle where a few broad white
flat scales may be present; psp scales absent. . . . . . .24. allotecnon
MALE GENITALIA

(18. lorraineae,; 25 guerrero; 26 guatemala; 28. marinkellei;
and 34. martinezi unknown; 35. vanemdeni not included)

Claspette completely absent. . . ta
Claspette represented by a small sternomesal lobe bearing a 1 differentiated
PO Se ee re ee ne ee Hee hie! Mee Le

Tergomesal and sternomesal margins of sidepiece with row of flattened foli-
form setae. soo 6.3 . . .22, whitmorei
Tergomesal and sternomesal - margins of sidepiece with simple bristles .
23. pseudodominicii

Apex of aedeagus in dorsal aspect uniformly sclerotized, smoothly rounded,
without indication of apical lateral lobes or apical horns .

ae pe § sexlineatus

Apex of Aedeaeus deeply lemareiriate and with distinicd long apical horns; or

dorsal sclerotization more or less emarginate apically, producing lateral

Sletten (ee ee a ee ee Or apa ae

Apex of aedeagus with slender horns, bent dorsad . 19. brevis; 20. spinosus
Dorsal sclerotization of aedeagus more or less deeply emarginate, producing
Riieral Serer Fes er eee

Spiniform of clasper short and thick, usually less than 0.25 of clasper length
Spiniform of clasper long and slender, usually at least 0.33 of clasper length

Aedeagus longer than proctiger ... sayy Catt wr duster iba atte oe
Aedeagus shorter than or subequal to proctiger Jenks ie amet staan veers Dis ai etem eT gO Se

Inner setae of ninth tergite lobe very markedly flattened, almost leaflike.
. . .d1. ecuadoriensis
Inner setae of ninth tergite lobe thickened but not markedly flattened . .8

Dorsal sclerotization of aedeagus deeply emarginate, lateral apical lobes con-
SPICUOUS “6 os . . 17. quadrivittatus
Dorsal sclerotization of aedeagus shallowly emarginate, lateral apical lobes in-
er Ra TR nd Oo Oe ane tee aoe

Spiniform of clasper about 0.5 of clasper length
bate & eleanorae. 33. durivittatus
Soiniforn of clasper jess than 0. 4 of clasper POG Make ce Chiat fuer es 2

Dorsal sclerotization of aedeagus deeply emarginate, lateral apical lobes con-
RCE a at Cate Wah area ii Mae Me tears

3(2).

4(2).

5(4).

6(5).

7(6).

8(6).

9(8).

2(1).

Berlin: Subgenus Howardina 3

Dorsal sclerotization of aedeagus shallowly emarginate, lateral lobes incon-
Sei cuss ua ke eae As Os ae er leat i 2 ga ee

PUPAE

(23. pseudodominicii; 26. guatemala; 31. ecuadoriensis; 32.
brevivittatus; 33. aurivittatus; 34. martinezi and 35. vanemdeni
unknown; 18. lorraineae and 25. guerrero not included)

Trumpet short and broad, index about3.0. . . . . . . .22. whitmorei
Trumpet dong and narrow tides over 3.0. ge Oe ee
Hairs 10,11,12-C all single . . diag 4 Pi eat i Sig ig a
At least one of these hairs branched, usually 10-C. Bae ey ai apt os at ea Ri hae
Hair 8-Vi) ondersal surfacts 2 hc ere i ec A ee ie ae eal
Hair 8-Vil on ventral surface. Sie 2 ies A ee Ora amelle
Hair 8-VII on dorsal surface, strongly developed. . . . . . 27. eleanorae
Hay B-V il ow ventral gorivce weak. ae i a Oe ee ee ee
Paddle-deeply emarinate-at gir bo. eee eee ee AS Brews
Paddle rounded or produced at heir a i2 0) ih Oa A OE er
Hair 5-IV-V moderate, not exceeding apex of second tergite following . .7
Hair 5-IV-V very long, exceeding apex of second tergite following . . . .8

Hair 9-VII weak and short, barely reaching lateral angle of tergite

. 30. osornoi

Hair 9-VIL stronger and longer, extending beyond lateral angle of tergite .
21. sexlineatus

Hair 1-II usually double or triple; 9-VIII with 3,4 branches .
fa ll ane allotecnon
Hair 1 TI usually at least 5-branched: 9-VIII with at least 4 branches . .9

Hair 9-VIII usually at least 8-branched; 9-VII at least 4-branched .
ety oie it quadrivittatus
Hair 9-VIII usually 6-branched: 9-VII double He eg gue As eo eeEES
LARVAE

(23. pseudodominicii; 26 guatemala; 31. ecuadoriensis;
32. brevivittatus; 33. aurivittatus; 34. martinezi and
35. vanemdeni unknown)

Ventral brush with 6 pairs of hairs ibn resi si ea ae
Ventral brush with 5 pairs of hairs . . . TSE ON
Quadrivittatus Group

Siphon without spicules visible at 10OX; hair 12-C double . . . ... .3

76

32).

4(2).

5(1).

6(5).

7(6).

8(7).

9(6).

10(9).

Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Siphon distinctly spiculate at 100X; hair 12-C stellate, with many branches

.4
Hair 5-C usually with 5 branches 2... 2 7 geedriviffatus
Hait 5-C usually with 3.4 branches .. ........ 4. ...c .. ..d6. lorraineae

Hair 5-C usually with more than 10 branches (11-22); ae index about

OS sa, “PhD: reves
Hair 5-C usually s branched (4- a. siphon index about 2: ie . 20. Spinosus
Whitmorei Group
Siphon with distinct spicules visible at 1OOX . . . . . . .22. whitmorei
Sip) Ort Without spices Vidivic gd) 10UX a ee
Hair 15-C usually triple, very short, at most Sait: level of base of mental
Diate. al
Hair 15-C usually single or r double, moderate i in leneth, reaching level of mid-
die Of mental pate (Eicanorde Group). se

Sexlineatus Group

Thoracic and abdominal integument with dense vestiture of hairlike spicules
oo 21, sexlineatus

Thoracic integument slibrous at 100X, abdominal with restricted dorsal and
ventral patch of small spinelike spicules (Allotecnon Group) .... .8

Allotecnon Group

Anal saddle with distinct large caudal marginal spicules ventrad of hair 1-X.
. .24. allotecnon
Anal saddle without any caudal marginal spicules ventrad of hair 1-X .
2), guerrero

Eleanorae Group

Siphonal hair 1-S within last pecten tooth; hair 5-C with more than 5
branches <.*~. . . 2/7, eleanorae
Siphonal hair 1-S distad of last pecten tooth: hair 5-C single or double .° 10

Peete a ee ie a esome,
Tinie oe odors 8 ee A a CO a a

11(10).Siphonal hair 1-S usually single and strongly barbed, rarely double; stellate

hairs all shorter than saddle Spee abdominal hairs 1 and 2 usually with
about 10-15 branches . . . . 28. marinkellei
Siphonal hair 1-S usually triple or - double: many stellate hairs distinctly long-
er than saddle, abdominal hairs 1 and 2 usually with 20 or more branches —
29. leei

Quadrivittatus Group

Berlin: Subgenus Howardina | qi

FEMALES. Head: Median light line on vertex with narrow scales. Proboscis dis-
tinctly longer than forefemur. Torus with a restricted mesal scale patch. Thorax:
Mesonotal light lines usually creamy to golden; inner dorsocentral line complete;
outer dorsocentral line either complete or incomplete, not reaching anterior margin;
lateral marginal line continuous; scales on anterior promontory golden. Antealar area
above paratergite with golden scales. Paratergite with a moderate scale patch extend-
ing entire sclerite length. Upper ppn entirely covered with light scales; upper stp
patch quadrate. Legs: Distal white spot on anterior surface of midfemur and hindfe-
mur absent. Tibiae completely dark. Wing: Veins entirely dark scaled. Abdomen:
Basomedian tergal light markings absent.

MALES. Palpus entirely dark scaled.

MALE GENITALIA. Lobe of tergite IX prominent to indistinct, bearing few (3-5)
strong or many (8-10) moderate bristles. Sidepieces separated at base by a distance
less than width of aedeagus; without any indication of lobes on tergomesal or sterno-
mesal margins. Claspette represented by a conical tubercle. Aedeagus shorter than
proctiger; apex moderate to deeply emarginate, producing lateral lobes or horns.

PUPAE. Cephalothorax: Trumpet moderate, index about 5.5-6.0, not expanded
apically; hair 10-C at least 5-branched; 12-C single, rarely double. Abdomen: Hair 5-
IV-V moderate to long, usually exceeding apex of second tergite following (except
in brevis); hair 9-VII distinctly stronger than hair 6, exceeding lateral angle of tergite.
Paddle: Moderate to long; apex usually produced (emarginate in brevis).

LARVAE. Head: Hairs 4-7-C moderately pigmented, long, all at least 5-branched;
hair 15-C long, exceeding apex of mental plate. Thorax: Roundish in outline, integu-
ment densely spiculate, spicules hairlike, simple or branched. Abdomen: Integument
strongly uniformly spiculate. Hair 8-VII slightly cephalodorsad of 9-VII. Siphon: In-
tegument imbricate or spiculate; acus distinct, separate. Anal Segment: Hair 3-X sin-
gle; ventral brush with 6 pairs of hairs.

DISCUSSION. The Quadrivittatus Group of 4 species is distinguished by unique
features both in the adults (absence of midfemoral and hindfemoral light spots) and
in the larvae (ventral brush with 6 pairs of hairs). Since both these characters are
primitive ones in the subgenus, this group probably represents the most ancient lin-
eage in the section. )

The group is known only in Central America from southern Mexico to central
Panama, usually at elevations above 4,000 feet, and is sympatric with the Allotecnon
Group (fig.5). A. quadrivittatus is the only widespread species, occupying the entire
known range of the group; lorraineae is restricted to the Sierra Madre del Sur and
Sierra Madre de Chiapas in Mexico, where it is sympatric with quadrivittatus, spino-
sus and brevis are sympatric with quadrivittatus in the southern part of the range,
spinosus in the mountains of Costa Rica and western Panama, and brevis apparently
only in western Panama.

17. Aedes (Howardina) quadrivittatus (Coquillett)
Figs.5 ,8,36,37
1902.Culex quadrivittatus Coquillett, 1902:292. TYPE: Holotype ¢, Chacula (Huchastenange).
Guatemala, elev. 6600 ft, 13 June 1902, Gustav Eisen [USNM, 6560].

Aedes (Howardina) quadrivittatus of Bonne and Bonne-Wepster (1925:369); Carpenter, Galindo
and Trapido (1952:163); Horsfall (1955:471); Trapido, Galindo and Carpenter (1955:530,537,

78 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

538); Vargas (1956:22); Trapido and Galindo (1957:122,125 ,130); Galindo (1964:34,56); Bel-
kin, Schick and Heinemann (1965:28); Forattini (1965 :401-403); Diaz Najera (1966b:63).

Aedes quadrivittatus of Dyar and Knab (1907:11); Theobald (1910:486); Picado (1913:353);
Howard, Dyar and Knab (1917:852-853); Dyar (1918:73); Kumm, Komp and Ruiz (1940:400,
417); Arnett (1949:238); Galindo, Carpenter and Trapido (1955:163); Stone and Knight (1956:
224); Blanton and Peyton (1958:31).

Aedes (Howardina) quadrivittatus in part of Dyar (1928:229-230); Edwards (1932:155); Lane
(1939:100; 1953:40); Stone, Knight and Starcke (1959:174).

Culex quadrivittatus of Blanchard (1905:628); Theobald (1905:29; 1910:395).

FEMALE. Wing: 3.33 mm. Proboscis: 2.16 mm. Forefemur: 1.8 mm. Abdomen:
about 3.06 mm. Head: Light decumbent scales golden on median stripe and short or-
bital line; sides and venter creamy. Erect scales on occiput dull golden. Palpus about
0.2 of proboscis, with a few white scales apically and mesally on segment 4. Basal
flagellar segment with dark scales. Thorax (fig.8): Mesonotal light scales golden; outer
dorsocentral line complete, only slightly expanded at fossa, merging with lateral mar-
ginal line at level of humeral bristles; lateral marginal line long, extending from hu-
meral bristles to wingbase. Median scutellar lobe with broad dark scales laterally and
narrow median golden scales from base to apex; lateral lobe with a few linear golden
scales. Paratergite with pale golden linear scales. Apn with narrow anterior and broad-
er posterior pale golden scales; ppn with only an upper patch of pale golden linear
scales; psp patch pale golden, linear; ssp patch with creamy broad scales; other pleu-
ral patches white, scales broad with distribution as in the group. Legs: Forefemur
and midfemur predominantly dark anteriorly except for a few basal creamy scales,
proximal half of posterior surfaces crearny and distal dark; hindfemur creamy up to
0.67 on both sides, rest dark. Foretarsus with a few basal white scales on segments
1,2; midtarsus with short, often incomplete basal white rings on 1,2; hindtarsus with
moderate white rings on 1-3; rest of tarsi dark. Haltere: Stem pale, knob with dark
scales basally; fringe whitish. Abdomen: Sternites with broad, V-shaped basal creamy
patch and rest dark; with a few apical creamy scales on II-V.

MALE. Wing: 3.42 mm. Proboscis: 2.43 mm. Forefemur: 1.98 mm. Essentially
similar to female in color. Submedian dark area of head narrower than in female.
Median scutellar lobe with larger number of golden scales.

MALE GENITALIA (fig.36). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with about 3-5 strong bristles. Sidepiece length 3.0 of medi-
an width. Apical seta of claspette on a short basal tubercle, broader than long.
Clasper about 0.6 of sidepiece, with 1,2 subapical hairs; spiniform short, about 0.25
of clasper. Aedeagus shorter than proctiger; dorsal sclerotization more or less deeply
emarginate, producing distinct lateral apical lobes.

PUPA (fig.36). Abdomen: 3.56 mm. Trumpet: 0.51 mm; index about o3Se3 Pad
dle: 0.81 mm. Readily separated from brevis by the elongate paddle with produced
apex; from spinosus by hair 9-VIII usually at least 8-branched. Chaetotaxy as fig-
ured, hairs moderately to darkly pigmented. Cephalothorax: Mesonotum between
trumpets moderately pigmented, lateral areas deeper; metanotum strongly pigmen-
ted; trumpet strongly contrasting. Hair 1(1, 1-3), 2(1, 1-2), 3(1), 4@, 2-3), 5@),
6(1), 7(2), 8(2, 1-3), 9(1), 10(6, 2-7), 11(1, 1-2), 12(1, 1-3). Abdomen: Tergite I
strongly pigmented except for usual membranous areas; segments II-IV strongly pig-
mented and imbricate, rest progressively lighter caudad. Segment I: hair 1(8-13 pri-
mary branches, each with several secondary branches), 2(1, 1-6), 3(1), 4(2, 2-3), 5(2,
1-3), 6(1, 1-2), 7(1, 1-2), 9(1). Segment II: hair 0(1), 1(5, 2-10), 2(1, 1-3), 3(1), 4G,
2-5), 51, 1-2), 6(1), 7C1, 1-2), 9(1). Segment III: hair 0(1), 1(1, 1-3), 2(1, 1-2), 3(1),

Berlin: Subgenus Howardina fie

4(1, 1-2), 5(1, 1-2), 6(1), 7(2, 1-3), 8(2, 1-3), 9(1), 10(1), 11(1), 14(1). Segment IV:
hair O(1), 1€1, 1-2), 21, 1-2), 3(2), 4(1, 1-2), 5(2), 6(1), 7(1, 1-3), 8(2, 1-3), 9(1),
10(1), 11¢€1), 14(1). Segment V: hair 0(1), 1(1, 1-2), 2(1), 3(1), 41, 1-2), 5(2, 1-2),
6(1), 7(2, 2-4), 8(2, 1-2), 9(1), 10(1), 11(1), 14(1). Segment VI: hair 0(1), 1(1),
2(1), 31), 41), 51, 1-2), 6(1, 1-3), 701, 1-2), 8(2, 1-2), 9(1), 10(1), 1101), 14(1).
Segment VII: hair 0(1), 1(1), 2(1), 3(1, 1-2), 4(1), 51), 6(1, 1-2), 7(1), 81, 1-2),
9(4, 3-6), 10(1), 11(1), 14(1). Segment VIII: hair 0(1), 4(1), 9(9, 7-11), 14(1, 1-2).
Paddle: Moderately to deeply pigmented, midrib darker; apex distinctly produced;
external margin serrate; hair 1 slender. Male genital lobe extending to 0.4 of paddle,
female genital lobe to 0.31.

LARVA (fig.37). Head: 0.86 mm. Siphon: 0.97 mm. Anal Saddle: 0.27 mm. Read-
ily separated from spinosus and brevis by the siphon without distinct spicules and
hair 12-C double; from lorraineae by hair 5-C usually 5-branched. Chaetotaxy as fig-
ured, hairs moderately to darkly pigmented. Head: Width subequal to length; pig-
mentation moderate except around mouthparts where it is deeper. Mental plate with
a strong median tooth and 11 distinct teeth (11-12) on each side, often a smaller de-
tached tooth at base. Hair 0(1), 1(1), 3(1), 4(8, 8-11), 5(6, 5-7), 6(7, 5-10), 79, 8-
15), 8(1), 965, 4-6), 10(2), 11°12, 10-17), 12(2, 2-4), 13(1), 14(12, 11-17), 15@3, 2-
3). Antenna: Length about 0.45 of head; pigmentation moderate. All hairs single ex-
cept 1-A which is branched (3, 2-3), and at 0.5 from base. Thorax: Integument with
simple or branched hairlike spicules. Stellate hairs with many barbed branches; long
hairs barbed and with attenuate apices. Prothorax: hair 0(19, 13-26), 1(5, 5-8), 2(1),
3(9, 8-11), 4(3, 3-4), 5(2, 1-3), 6(1), 7(2), 8(16, 16-20), 9(9, 7-10), 10(1), 112, 1-3),
12(1), 13(24, 20-27), 14(38, 38-48). Mesothorax: hair 1(11, 10-16), 2(1), 3(1), 4(1),
5(1), 6(6, 5-6), 7(1), 8(8, 6-9), 9(9, 7-10), 10(1), 11(1), 121), 13(6, 3-8), 14(8, 7-
12). Metathorax: hair 1(14, 11-18), 2(1), 3(4, 3-7), 4(15, 12-20), 5(15, 13-20), 6(1),
7(8, 7-10), 8(7, 4-8), 9(5, 5-7), 10(1), 11€1), 12(1), 1317, 16-23). Abdomen: Integ-
ument and stellate hairs as on thorax. Segment I: hair 1(21, 16-25), 2(21, 19-28),
3(1), 4(1), 5(14, 14-19), 6(3, 3-4), 7(1), 9(4, 3-6), 10(1), 11(18, 16-24), 13(10, 8-
17). Segment II: hair 0(1), 1(18, 17-22), 2(22, 17-28), 3(1), 4(1), 5(18, 16-21), 6(3,
3-5), 7(8, 7-9), 8(1), 9(6, 5-8), 10(1), 11(1), 12(1), 1314, 12-19). Segment III: hair
O(1), 1(21, 18-25), 2(18, 16-26), 3(1), 4(1), 5(21, 16-26), 6(1), 7(10, 8-11), 8(1),
9(12, 7-15), 10(1), 11(1), 12(1), 13(20, 17-21), 14(1). Segment IV: hair 0(1), 1(20,
18-27), 2(17, 16-27), 3(1), 4(1), 5(21, 15-24), 6(1), 7(8, 8-10), 8(1), 9(16, 10-16),
10(1), 11(€1), 12€1), 13(20, 15-21), 14(1). Segment V: hair 0(1), 1(21, 19-27), 2(16,
16-22), 3(1), 4(1), 5(20, 14-20), 6(1), 7(9, 8-12), 8(1), 9(16, 11-19), 10(1), 11(1),
12(1), 13(18, 14-20), 14(1). Segment VI: hair 0(1), 1(17, 17-23), 2(18, 14-21), 3(1),
4(1), 5(17, 15-19), 6(1), 7(9, 8-13), 8(1), 9(15, 10-16), 10(1), 11(1), 12(1), 1315,
11-15), 141). Segment VII: hair 0(1), 1(18, 16-21), 2(16, 14-19), 3(1), 4(1), 5(16,
9-19), 6(6, 4-7), 7(2, 2-3), 8(2, 1-3), 9(9, 8-14), 10(1), 11(1), 12(1), 13(4, 3-5), 14(1).
Segment VIII: Comb scales 16(14-20), deeply pigmented, elongate; lateral fringes —
not extending to apex. Hair 0(1), 1(11, 9-14), 2(1), 3(8, 6-10), 4(1), 5(5, 5-6), 14(1).
Siphon: Index about 4.0-4.2; integument imbricate, moderately to deeply pigmen-
ted, with darker basal ring. Pecten with 17(13-18) teeth extending to 0.4 and termi-
nating before hair 1-S; individual tooth long, pointed, with basal denticles. Hair 1(3,
3-4), 2(1), 3-5 not visible, 6(1), 7(1), 8(2, 1-2), 9(1). Anal Segment: Saddle deeply
pigmented and imbricate; caudal margin with spines of varying length, long ones pre-
dominating. Gills long, subequal, as long as ventral brush. Hair 1(2), 2(5, 4-5), 3(1),
4a(4, 3-4), 4b(3, 2-3), 4c(3, 2-4), 4d(3, 2-4), 4e(3, 3-4), 4£(3, 2-4).

SYSTEMATICS. A. quadrivittatus is the only species occupying the entire range

80 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

of the Quadrivittatus Group from the southern end of the Sierra Madre Oriental in
Mexico to central Panama. I have seen no specimen of quadrivittatus from eastern
Panama and all material from South America, formerly reported as this species, actu-
ally belongs to new species in the Eleanorae Group. A. quadrivittatus is readily dif-
ferentiated in all stages from the 3 other members of the group which have restricted
distributions, all peripherally sympatric with quadrivittatus. Its closest relative is un-
doubtedly lorraineae (see). Very little variation is apparent throughout the wide
known distribution of quadrivittatus except at the northern end in Mexico where
the light thoracic scaling of the adults tends to be more yellowish or creamy. No sig-
nificant population differentiation in the immature stages has been detected in the
relatively large sample examined.

BIONOMICS. A. quadrivittatus apparently breeds only in epiphytic bromeliads,
primarily in cloud forest areas at elevations above 4,000 feet. Females are readily at-
tracted to human bait and display a distinct preference for feeding in the forest cano-
py (Carpenter, Galindo and Trapido 1952:163). Recently an unidentified arbovirus
has been isolated from quadrivittatus in Panama (Galindo 1964:34).

DISTRIBUTION (fig.5). Cordilleras of Middle America in Mexico (southern part
of Sierra Madre Oriental and eastern end of Sierra Madre del Sur), Guatemala, Hon-
duras, El Salvador, probably Nicaragua, Costa Rica and Panama (western part) at
elevations varying from about 4000 ft to 8400 ft. Material examined: 321 specimens;
15 6, 42 9, 43 pupae, 221 larvae; 31 individual rearings (6 pupal, 13 larval, 12 incom-
plete).

COSTA RICA. Alajuela: Poasito, 14 Nov 1962, C.L. Hogue and W.A. Powder (CR 23), 1 Ip?
(23-105), 1 L [UCLA]. Volcan Poas, 1 p? (156), 1 ? [USNM] ; Summit, 9 July 1963, C.L. Hogue,
1 L(CR 145) [UCLA]. Cartago: Elev. 1500 m, C. Picado, 2 9? [USNM]. Cervantes, 9 Nov 1962,
J.N. Belkin, C.L. Hogue and W.A. Powder, 3 L (CR 13) [UCLA]. Cipreses, 4 6, 29,9 L, 31
[USNM]. La Sierra, elev. 6300-8400 ft, 24 Nov 1962, C.L. Hogue and W.A. Powder, 1 IP (49-107),
1 6,7 L (CR 49), 1 Ipé (51-105), 1 lp? (51-103), 2 IP (51-107,110), 23,6 P, 1 p, 25 L(CR 51),
1 Ipé (52-105), 2 IP (52-106,108), 12 L(CR 52), 3 Ipd (53-101-103), 1 P, 2 L(CR 53), 1 Ipd (54-
103), 1 Ip? (54-106), 5 IP (54-101,102,104,105,107), 3 P, 1 p, 117 L(CR 54) [UCLA].

EL SALVADOR. Los Esesmiles, H W. Kum, 2k [USNM].

GUATEMALA. Chimaltenango: San Andres Semetabaj, 17 July 1965, iCowsill (GUA 50), 1 lp?
(50-10) [UCLA]. Solola, 30 June 1964, J.N. Belkin and J.E. Zavortink (GUA 15), 2 Ipd (15-10,
12), 1 Ip? (15-11), 2 1P (15.13.14), 3 L [UCLA] . Huehuetenango: Chacula, elev. 6600 ft, 13 June
A a Eisen, type series, 6 2 [USNM] . Locality not specified: Sept 1902, G. Eisen, 1 9; 14,39

HONDURAS. Tegucigalpa, 17 July 1945, 1 lp [USNM].

MEXICO, Chiapas: Paval, 28 A (Diaz Najera 1966b:61). Teopisca, 19 Aug 1964, E. Fisher and
D. Verity, 10 L (MEX 132) [UCLA]. Triunfo, 125 A (Diaz Najera 1966b:61). Morelos: (Vargas
1956:22). Oaxaca: Ixtlan de Juarez, Vivero Rancho Teja, elev. 7200 ft, 12 Aug 1966, D.A. Schroe-
der (MEX 430), 1 pd (430-105), 4 p? (430-101-104), 12 9 (MEX 433) [UCLA]. Tamaulipas: Ran-
cho del Cielo, 4 Apr 1963, Trevino, 2 2 [USNM]. Veracruz: Banderilla, Feb 1945, G. Lissmann,
2g [UCLA].

PANAMA. Bocas del Toro: near Almirante, 1952-53, 2297 ? (Trapido and Galindo 1957:122,
125). La Zorra, elev. 4400 ft, 5 Apr 1963, A. Quinonez, 2 ? (PA 174), 1 9? (PA 175) [UCLA].
Panama: Cerro Campana, 28 Aug 1963, A. Quinonez, 1 9 (PA 537) [UCLA].

18. Aedes (Howardina) lorraineae Berlin, n.sp.
Figs.5 ,8,38

Berlin: Subgenus Howardina 81

TYPES: Holotype ? (MEX 433), Vivero Rancho Teja, Ixtlan de Juarez, Oaxaca, Mexico, 7200
ft, biting collection, 12 Aug 1966, D.A. Schroeder [USNM]. Paratypes: 7 2 (MEX 433), same data
as holotype [USNM, BM, UCLA]. This species is dedicated to the memory of Mrs. Lorraine Belkin.

FEMALE. Wing: 3.85 mm. Proboscis: 2.62 mm. Forefemur: 2.19 mm. Abdomen:
about 3.15 mm. Generally similar to quadrivittatus and other members of the group,
differing primarily in the following. Head: Light decumbent scales on median stripe
and short orbital line creamy. Erect scales creamy. Palpus about 0.2 of proboscis,
with a few white scales on mesal surface of segment 4. Basal flagellar segment with a
few white scales on dorsomesal surface. Thorax (fig.8): Mesonotal light lines creamy
to pale golden; outer dorsocentral line complete, broadly joined anteriorly in fossa to
entire lateral prescutal area and reaching to scutellum; lateral marginal line shorter,
extending from scutal angle to wingbase. Narrow light scales on median scutellar lobe
creamy; lateral lobes with a few narrow creamy scales. Paratergite and psp with mod-
erately broad, white scales; apn patch creamy; upper ppn with predominantly broad
creamy scales; lower ppn white, adjoining ssp patch; other pleural patches white, dis-
tributed as in the group. Legs: Forefemur and midfemur predominantly dark anteri-
orly except for scattered creamy scales up to proximal 0.5, proximal 0.5 of posterior
surfaces creamy, extending distally as narrow line, rest dark; hindfemur creamy prox-
imally to 0.6 on both sides and rest dark. Foretarsus with basal white scales on seg-
ment 1; midtarsus with short incomplete white rings on 1,2; hindtarsus with moder-
ately broad white rings on 1-3. Haltere: Stem pale, knob predominantly white, with
a few basal dark scales. Abdomen: Sternites with broad V-shaped basal white patches;
apical white scales numerous on II-IV, few on V and absent on distal segments.

MALE unknown and PUPA imperfectly known.

LARVA (fig.38). Head: 0.81 mm. Siphon: 0.87 mm. Anal Saddle: 0.27 mm. Diag-
nostic characters as in the key. Readily separated from quadrivittatus by hair 5-C
usually only 3,4-branched; from brevis and spinosus by the absence of spicules on
siphon. Head: Mental plate with a strong median tooth and 11 distinct teeth (10-12)
on each side, often a small detached tooth at base. Hair 0(1), 1(1), 3(1), 4(7, 5-11),
5(3, 3-4), 6(5, 5-6), 7(6, 6-9), 8(1), 9(2, 1-3), 10(2), 1111, 9-17), 12(@, 1-2), 13(1),
14(8, 6-13), 15(2, 2-3). Antenna: Length about 0.33 of head; uniformly moderately
pigmented, with a few scattered spicules; hairs simple except 1(2, 2-3). Thorax: In-
tegument predominantly with simple hairlike spicules. Stellate hairs with many un-
equal barbed branches; long hairs barbed and with attenuate apices. Prothorax: hair
O(16, 13-17), 1(6, 4-6), 2(1), 3(8, 6-10), 4(4, 2-4), 5(2), 6(1), 7(2), 816, 14-19),
9(8, 7-9), 10(1), 11@, 1-2), 12(1), 13(@22, 19-27), 14(40, 38-56). Mesothorax: hair
1(11, 10-18), 2(1), 31), 4(€1), 51), 6(6, 4-6), 7(1), 8(7, 6-8), 9(6, 6-9), 10(1), 11d,
1-2), 12(1), 13(6, 5-9), 14(7, 6-9). Metathorax: hair 1(13, 10-18), 2(1), 3(4, 4-6),
4(10, 10-17), 5(13, 8-16), 6(1), 7(7, 4-8), 8(6, 6-9), 9(5, 4-7), 101), 11€1, 1-2),
12(1), 13@20, 14-22). Abdomen: Long and stellate hairs as in thorax. Segment I: hair
1(18, 15-21), 2(16, 14-22), 3(1), 4(1), 5(12, 10-16), 6(3, 2-3), 71), 9(5, 4-7), 101),
11(17, 14-21), 13(15, 9-21). Segment II: hair 0(1), 1(17, 13-20), 2(20, 12-23), 3(1),
4(1, 1-2), 5(16, 13-19), 6(3), 7(8, 5-10), 8(1), 9(6, 6-9), 10(1), 11(€1), 12(1), 13(14,
11-17). Segment III: hair 0(1), 1(20, 15-22), 2(16, 14-21), 3(1), 4(€1), 5(16, 13-20),
6(1, 1-2), 7(10, 8-11), 8(1), 9(10, 8-16), 10€1), 11(1), 121), 13(16, 16-19), 14(1).
Segment IV: hair 0(1), 1(20, 15-23), 2(19, 15-19), 3(1), 4(1), 5(16, 14-22), 6(1, 1-2),
7(10, 7-12), 8(1), 9(12, 10-18), 10(1), 11¢€1), 121), 13(17, 13-21), 14(1). Segment
V: hair 0(1), 1(19, 14-21), 2(18, 12-18), 3(1), 4(1), 5(16, 11-17), 6(1, 1-2), 7(11, 8-
11), 8(1), 9(12, 10-17), 10€1), 11(1), 12(1), 13(15, 12-19), 14(1). Segment VI: hair

82 | Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

O(1), 1018, 12-23), 2(17, 12-18), 3C1), 4(1), 5(16, 12-17), 6(1, 1-2), 7010, 8-12),
8(1), 9(13, 8-18), 10(1), 11€1), 12(1), 1312, 10-19), 14(1). Segment VII: hair 0(1),
1(18, 13-23), 2(17, 10-18), 3(1), 4(1), 5(16, 13-20), 6(4, 3-5), 7(2, 2-3), 8(2, 2-4),
9(8, 6-11), 10(1), 11(1), 121), 13(4, 3-4), 14C1). Segment VIIT: Comb scales usually
16(14-20), moderately pigmented, their lateral fringes not extending to apex. Hair
O(1), 1(8, 7-16), 2(1), 3(8, 7-9), 41), 5(6, 5-8), 14(1, 1-2). Siphon: Index 3.3-3.6;
integument imbricate, moderately pigmented, with darker basal ring. Pecten teeth
usually 16(13-18), extending to 0.4. Hair 1(4, 3-5), 2(1), 3-5 not visible, 6(1), 7(1),
8(2, 1-2), 9(1). Anal Segment: Saddle moderately pigmented. Gills subequal, narrow,
as long as ventral brush. Hair 1(2, 2-3), 2(5, 4-6), 3(1), 4a(3, 3-4), 4b(3, 2-3), 4c(3,
2-3), 4d(3, 2-3), 4e(3, 2-3), 4f(3, 2-3).

SYSTEMATICS. A. lorraineae is undoubtedly closely related to quadrivittatus. It
is differentiated in the female most conspicuously by the white scales on the first fla-
gellar segment and generally whiter scales on the mesonotum and pleura. The associ-
ation of the immature stages with the females is presumptive only since all the rear-
ings are incomplete. The larva described here, although very similar to quadrivittatus,
is distinct from those of Mexican and other populations of the latter in the smaller
number of branches in head hair 5. The pupal stage is also in general similar to quad-
rivittatus but cannot be described at this time since only a few imperfect whole pu-
pae from incomplete rearings are available. A. lorraineae is known at present only
from the Sierra Madre del Sur and the Sierra Madre de Chiapas in Mexico at eleva-
tions above 7,000 feet. In the eastern part of this range, in the states of Oaxaca and
Chiapas, lorraineae is sympatric with quadrivittatus. In one biting-landing collection
(MEX 433) in Oaxaca where the 2 species were taken in approximately equal num-
bers there is no indication of intergradation between lorraineae and quadrivittatus.
This supports my interpretation that lorraineae is a distinct species and not a geo-
graphical isolate of quadrivittatus.

BIONOMICS. Immature stages of lorraineae have been collected in epiphytic bro-
meliads only. Females of this species are apparently readily attracted to humans.

DISTRIBUTION (fig.5). Sierra Madre del Sur and Sierra Madre de Chiapas in Mex-
ico at elevations above 7000 ft. Material examined: 75 specimens; 11 °, 5 pupae, 59
larvae; 5 incomplete individual rearings.

MEXICO. Chiapas: 27 Mar 1961, R.B. Eads, 1 L [USNM]. Guerrero: Omilteme, 8 Sept 1965,
D.A. Schroeder, 3 9 (MEX 361A), 3 IP (362-21-23), 24 L (MEX 362) [UCLA]. Oaxaca: Ixtlan de
Juarez, El Punto, elev. 7700 ft, 13 Aug 1966, D.A. Schroeder, 4 L (MEX 434); Vivero Rancho
Teja, elev. 7200 ft, 12 Aug 1966, D.A. Schroeder, 2 IP (428-10,11), 9 L (MEX 428), 4 L (MEX
429), 9 L (MEX 430), 3 L (MEX 431), 8 9 (MEX 433, type series) [USNM, BM, UCLA].

19. Aedes (Howardina) brevis Berlin, n.sp.
Figs.5 ,9 39,40

TYPES: Holotype ¢d (PA 171-104) with associated pupal skin and genitalia slide, La Zorra,
Chiriquisito, Chiriqui Grande, Bocas del Toro, Panama, 4400 ft, 5 Apr 1963, A. Quinonez [USNM].
Allotype ? (PA 171-102) with associated pupal skin, same data as holotype [USNM]. Paratypes:
1 19 (171-105), 2 IP (171-101,103), 2 P, 18 L, same data as holotype [UCLA].

FEMALE. Wing: 3.78 mm. Proboscis: 2.7 mm. Forefemur: 2.34 mm. Abdomen:
about 3.1 mm. Generally similar to quadrivittatus and other members of the group;
differing primarily in the following. Head: Palpus about 0.15 of proboscis; with dis-

Berlin: Subgenus Howardina 83

tinct apical white scales on segment 4. Thorax (fig.9): Outer dorsocentral line incom-
plete, arising at anterior 0.33 and reaching to scutellum; lateral marginal line long,
extending from anterior promontory to wingbase, slightly curved inwards at lateral
prescutal area. Lateral scutellar lobes with a few narrow dark scales. Paratergite and
psp with narrow creamy scales; other pleural scale patches as in quadrivittatus, all
dull white. Legs: Forefemur dark anteriorly except for a few basal dingy scales, prox-
imal 0.5 of posterior surface dingy white, extending distally as narrow line bordered
by dark scales; midfemur dark anteriorly, proximal 0.6 of posterior surface dingy
white, distal dark; hindfemur dingy white proximally to 0.67 on both sides, rest
dark. Foretarsus with basal white scales on segments 1,2; midtarsus with short in-
complete white rings on 1,2; hindtarsus with broader white rings on 1-3, last covering
about 0.65 of segment. Haltere: Stem pale; knob predominantly white, with a few
dorsal dark scales. Abdomen: Sternites with broad V-shaped basal creamy patches;
apical light scales apparent on II, III.

MALE. Wing: 3.33 mm. Proboscis: 2.89 mm. Forefemur: 2.28 mm. Essentially
similar to the female in color.

MALE GENITALIA (fig.39). As figured; diagnostic characters as in the key. Ninth
tergite lobe indistinct, with 8-10 moderate bristles. Sidepiece length about 3.0 of me-
dian width. Clasper about 0.5 of sidepiece, with 1,2 subapical hairs; spiniform short,
length about 0.25 of clasper. Apex of aedeagus with slender horns bent dorsad.

PUPA (fig.39). Abdomen: 3.78 mm. Trumpet: 0.54 mm; index about 6.0. Paddle:
0.7 mm. Readily separated from other species of group by the deeply emarginate
paddle. Chaetotaxy as given below, hairs moderately to darkly pigmented. Cephalo-
thorax: Mesonotum lightly pigmented between trumpets, lateral areas darker; meta-
notum moderately pigmented. Hair 1(1, 1-2), 2(1), 3C1), 4(2), 5(2), 6(1), 7(2, 1-2),
8(2, 2-3), 9(1), 10(12, 12-14), 111), 12C1, 1-3). Abdomen: Tergite I uniformly mod-
erately pigmented except for usual membranous areas; other segments moderate, pig-
mentation progressively lighter caudad. Segment I: hair 1(10-20 primary branches,
each with secondary branches), 2(2, 1-4), 3(1), 4(3, 2-4), 5(3, 2-5), 6(1), 7(1, 1-2),
9(1); in some hair 10(1) present. Segment II: hair 0(1), 1(16, 12-20), 21, 1-2), 3(1),
4(3, 1-3), 5(2, 1-3), 6(1), 7(1), 9(1), 11(1). Segment III: hair 0(1), 1(2, 2-3), 2(1),
3(1), 41, 1-2), 5(1, 1-5), 6(1), 7€1, 1-2), 8(1), 91), 10(1), 11¢1), 14(1). Segment
IV: hair 0(1), 1(2), 2(1), 3(2, 1-2), 4(1), 5(2), 6(1), 701), 81), 91), 10(1), 1101),
14(1). Segment V: hair 0(1), 1(2, 1-2), 2(1), 3(1), 4(1), 5(2), 6(1), 7@, 1-3), 8(1),
9(1), 10(1), 11(1), 14(1). Segment VI: hair 0(1), 1(1, 1-2), 21), 301), 4(1), 5(),
6(1), 7(2, 1-6), 8(1), 9(1), 10€1), 1101), 14(1). Segment VII: hair 0(1), 1(1, 1-2),
2(1), 3(1), 4(1), 5(1), 6(2, 1-3), 7(1), 8C1, 1-2), 9(7, 5-8), 10(1), 11(1), 141). Seg-
ment VIII: hair 0(1), 4(1), 9(11, 11-14), 14(1). Paddle: Uniformly lightly pigmented,
midrib slightly darker; deeply emarginate; external margin serrate; hair 1-P single.
Male genital lobe extending to about 0.5 of saddle, female genital lobe to 0.35.

LARVA (fig.40). Head: 0.96 mm. Siphon: 0.6 mm. Anal Saddle: 0.3 mm. Readi-
ly separated from other members of the group by hair 5-C usually with 18 branches
(11-22) and short siphon with an index of about 2.0. Head: Mental plate with a
strong median tooth and 10 distinct teeth (9-10) on each side, often a small detached
tooth at base. Hair 0(1), 1(1), 3(1), 4(16, 13-20), 5(18, 11-22), 6(12, 9-19), 7(15,
~ 11-21), 8(4, 1-7), 9°09, 6-18), 10(6, 3-8), 11(24, 18-25), 12(8, 5-12), 13(1), 14Q19,
15-22). 15(2, 2-4). Antenna: Length about 0.35 of head; shaft uniformly moderate-
ly pigmented, with a few scattered spicules; hairs simple except 1(2, 2-3). Thorax:
Integument with simple and branched, hairlike spicules. Stellate hairs with many un-
equal barbed branches; long hairs barbed and with attenuate apices. Prothorax: hair

84 : Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

0(24, 19-24), 1(12, 10-14), 2(1, 1-2), 3(14, 13-23), 4(4, 3-6), 5@, 2-3), 6(1), 7(3, 2-
3), 8(21, 16-26), 9(7, 6-12), 10(1), 11(@2, 1-2), 121), 13(30, 24-31), 14(38, 34-43).
Mesothorax: hair 1(21, 15-33), 2(1), 31), 4(1), 5(1, 1-2), 6(6, 6-9), 7(1), 8(7, 5-7),
99, 7-10), 10(1), 11€1, 1-2), 12(1), 13(16, 9-16), 14(16, 15-21). Metathorax: hair
1(28, 22-30), 2(1), 3(16, 12-19), 4(5, 4-5), 5(12, 11-15), 6(1), 7(6, 4-9), 8(14, 10-
18), 9(5, 5-8), 10€1), 111), 12(1), 1319, 19-35). Abdomen: Stellate hairs as in tho-
rax. Segment I: hair 1(29, 29-32), 2(22, 18-31), 3(1), 41, 1-3), 5(18, 14-23), 6(3, 1-
4), 7(1), 9(7, 5-8), 10(1), 11022, 17-22), 13(26, 23-27). Segment II: hair 0(1), 1(26,
24-40), 2(22, 15-26), 3(1), 4€1), 5(18, 16-23), 6(3, 3-4), 7(13, 5-14), 8(1), 9(13, 9-
1S), 101), 1101), 121), 13(25, 21-26). Segment III: hair 0(1), 1(26, 24-35), 2(18,
15-22), 31), 4(1), 5(18, 17-22), 6(3, 2-3), 7014, 11-15), 8(1), 9(16, 12-16), 10(1),
11q1), 121), 13(21, 19-25), 14(1). Segment IV: hair 0(1), 1(26, 24-29), 2(16, 16-
20), 31), 41), 519, 15-21), 6(2), 711, 11-18), 8(1), 9(15, 15-18), 10(1), 1101),
12(1), 13(21, 20-34), 14(1). Segment V: hair 0(1), 1(25, 24-37), 2(18, 15-20), 3(1),
4(1), 5(18, 16-21), 6(2, 2-3), 7(12, 9-18), 8(1), 9(16, 14-24), 10(1), 11(1), 12(1),
13(19, 18-34), 14(1). Segment VI: hair 0(1), 1(27, 24-38), 2(18, 15-19), 3(1), 4(1),
5(16, 14-20), 6(2), 7(12, 11-14), 8(1), 9(16, 14-20), 10(1), 111), 1201), 13(20, 17-
22), 14(1). Segment VII: hair 0(1), 1(25, 23-30), 2(17, 14-22), 3(1), 4(1), 5(16, 14-
18), 6(7, 5-7), 72, 1-2), 8(3, 2-4), 9(19, 14-20), 10(1), 11(1), 12(1), 1319, 17-23),
14(1, 1-2). Segment VIIT: Comb scales usually 16(14-17), moderately pigmented,
their lateral fringes not extending to apex. Hair 0(1), 1(18, 15-21), 2(1), 3(7, 6-9),
4(1), 5(16, 15-20), 14(1). Siphon: Index about 2.0; integument moderately pigmen-
ted, with darker basal ring and with rows of small hairlike spicules. Pecten teeth usu-
ally 9(7-16), extending to 0.3. Hair 1(5, 4-6), 2(1), 3-5 not visible, 6(1), 7(1), 8(1, I-
2), 9(1). Anal Segment: Saddle moderately pigmented, with a darker basal ring and
with small spicules; caudal margin with spines of varying length. Gills about 2.0 of
saddle length, narrow and subequal. Hair 1(2, 2-3), 2(6, 4-6), 3(1), 4a(4, 3-5), 4b(3,
2-3), 4c(3, 3-4), 4d(4), 4e(4, 3-5), 4£(4, 3-5).

SYSTEMATICS. The adults of brevis cannot be differentiated from the apparent-
ly closely related spinosus but the immature stages of the 2 species are strikingly dif-
ferentiated as indicated in the diagnosis and in the keys. The association of all the
stages is undoubtedly correct although no complete larval rearings were made. At
present, brevis is known only from northern Panama where it has been found in the
same locality as quadrivittatus and spinosus but not in the same collections.

BIONOMICS. Immature stages of brevis have been collected in epiphytic bromeli-
ads at elevations above 4,000 feet. All 5 known adults were reared but it is possible
that the 2 females taken in biting-landing collections in Panama provisionally identi-
fied as spinosus are actually brevis.

DISTRIBUTION (fig.5). Chiriqui mountains of Panama at elevations above 4000
ft. Material examined: 58 specimens; 3 6, 2 2, 10 pupae, 43 larvae; 8 individual rear-
ings (1 larval, 4 pupal, 3 incomplete).

PANAMA. Bocas del Toro: La Zorra, elev. 4400 ft, 5 Apr 1963, A. Quinonez (PA 171, type se-
ries), 1 1d (171-105), 1 pd (171-104), 1 p? (171-102), 2 IP (171-101,103), 2 P, 18 L;6 Apr 1963,
A. Quinonez, 18 L (PA 178); 9 Apr 1963, A. Quinonez (PA 183), 1 pd (183-101), 1 p? (183-103),
1 IP (183-102), 3 L [USNM, UCLA].

20. Aedes (Howardina) spinosus Berlin, n.sp.
Figs.5 ,9,41,42

Berlin: Subgenus Howardina 85

TYPES: Holotype ? (CR 13-111) with associated pupal and larval skins, near Cervantes, Car-
tago, Costa Rica, 9 Nov 1962, J.N. Belkin, C.L. Hogue and W.A. Powder [USNM]. Paratypes: 4
larvae (CR 13), same data as holotype [USNM, UCLA].

FEMALE. Wing: 3.33 mm. Proboscis: 2.32 mm. Forefemur: 1.8 mm. Abdomen:
3.06 mm. Holotype lacking part of hindtarsus, entire left leg and abdomen. Essential-
ly similar to brevis in all details, from which it cannot be separated. The only appar-
ent difference is in the pleural scales which are white in spinosus and dingy white in
brevis (fig.9).

MALE. Wing: 3.5 mm. Proboscis: 2.8 mm. Forefemur: 2.15 mm. Essentially sim-
ilar to female in color.

MALE GENITALIA (fig.41). Diagnostic characters as in the key. Essentially simi-
lar to brevis in all details.

PUPA (fig.41). Abdomen: 3.25 mm. Trumpet: 0.5 mm; index about 6.0. Paddle:
0.77 mm. Readily separated from brevis by the rounded or slightly produced paddle;
from quadrivittatus by 2-branched hair 9-VII and 5-branched 9-VIII. Hairs moder-
ately to darkly pigmented. Cephalothorax: Mesonotum lightly pigmented between
trumpets, lateral areas and metanotum slightly darker; trumpets moderately long,
strongly contrasting. Hair 10-C with 5 branches; 12-C double. Abdomen: Tergite I
moderately pigmented except for usual membranous areas; pigmentation of rest of
segments moderate, progressively lighter caudad. Hair 1-II 8-branched; hair 5-IV,V
long, double, exceeding apex of second tergite following; hair 9-VII distinctly larger
than 6-VII, double, long, exceeding apex of tergite; hair 9-VIII long, 5-branched.
Paddle: Uniformly lightly pigmented, midrib slightly darker; apex rounded or slight-
ly produced; hair 1-P single; external margin serrate. Female genital lobe extending
to 0.38 of paddle.

LARVA (fig.42). Head: 0.88 mm. Siphon: 0.88 mm. Anal Saddle: 0.33 mm. Read-
ily separated from quadrivittatus and lorraineae by the pilose siphon; from brevis by
the longer siphon and hair 5-C usually 5-branched (4-7). Head: Mental plate with a
strong median tooth and 11 distinct teeth (11-12) on each side, often a small de-
tached tooth at base. Hair 0(1), 1(1), 3(1), 409, 8-11), 5(5, 4-7), 6(8, 6-9), 7(8, 8-11),
8(2, 1-3), 9(7, 5-9), 10(4, 3-5), 11(21, 17-24), 12(8, 4-10), 13(1), 14(16, 15-24),
15(3). Antenna: Length about 0.33 of head; shaft moderately pigmented and with a
few scattered spicules; hairs simple except 1(2, 2-3) which is at about 0.6 from base.
Thorax: Integument predominantly with branched, hairlike spicules. Stellate hairs
with many unequal barbed branches; long hairs with attenuate apices. Prothorax:
hair 0(18, 16-20), 1(5, 5-8), 2(1), 3(11, 10-14), 4(3, 3-5), 5(2), 6(1), 7(2), 8(16, 16-
20), 9(6, 4-7), 10(1), 11(2, 2-3), 12(1), 13(31, 20-38), 14(44, 44-60). Mesothorax:
hair 1(18, 9-20), 2(1), 3(1), 4(1), 5(1), 6(6, 4-6), 7(1), 8(7, 6-7), 9(7, 4-7), 10(1),
11(2), 121), 1365, 4-7), 14(8, 5-14). Metathorax: hair 1(25, 14-27), 2(1), 3(4, 3-4),
4(18, 14-20), 5(18, 15-20), 6(1), 7(7, 5-8), 8(10, 5-10), 9(5, 4-6), 10(1), 11(1, 1-2),
12(1), 13(20, 18-24). Abdomen: Segment I: hair 1(28, 24-33), 2(25, 22-27), 3(1),
4(1), 5(20, 16-23), 6(3), 7(1, 1-2), 9(6, 6-8), 10(1), 11(22, 19-25), 13(24, 20-28).
Segment II: hair 0(1), 1(30, 26-34), 2(24, 21-28), 3(1), 4(1), 5(28, 22-31), 6(3),
7(14, 10-14), 8(1), 9(14, 11-16), 10(1), 1101), 121), 13(29, 26-31). Segment III:
hair O(1), 1(30, 22-35), 2(25, 22-29), 3(1), 4(1), 5(28, 22-31), 6(2), 7(16, 10-16),
8(1), 9(20, 11-22), 10(1), 11(1), 12(1), 13(28, 18-32), 14(1). Segment IV: hair 0(1),
1(30, 21-36), 2(27, 20-28), 3(1), 4(1), 5(26, 19-31), 6(2), 7(16, 9-17), 8(1), 9(20,
I-22), LOT), 111), 12019, 13@7,; 20-32), 14€1). Segment V; hair 001), 1134, 23-
34), 2(24, 16-28), 3(1), 4(1), 5(26, 20-31), 6(2), 7(16, 9-19), 8(1), 9(21, 16-24),

86 : Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

10(1), 11(1), 12(1), 13(27, 18-28), 14(1). Segment VI: hair 0(1), 1(32, 21-33), 223,
17-27), 3(1), 4(1), 5(27, 19-29), 6(2), 7(17, 11-20), 8(1), 9(20, 15-24), 10(1), 11(),
12(1), 13(20, 16-24), 14(1). Segment VII: hair 0(1), 1(30, 22-34), 2(25, 17-26), 3(1),
4(1), 5€18, 17-24), 6, 6-11), 7(@, 2-3), 8C1), 9@2, 10-23), 10G); 11(1), 120),
13(17, 13-20), 14(1). Segment VIII: Comb scales usually 16(14-18), moderately pig-
mented; individual scale long, pointed apically and fringed nearly to tip. Hair O(1),
1(22, 14-26), 2(1), 3(8, 7-9), 4(1), 5(12, 8-16), 14(1). Siphon: Index about 3.2-4.0;
moderately pigmented, except for dark basal ring; uniformly covered with rows of
small hairlike spicules. Pecten teeth usually 10(10-15), extending to about 0.4 before
hair 1-S; individual tooth long, pointed, with basal denticles. Hair 1(5, 4-5), 2(1), 3-5
not visible, 6(1), 7(1), 8(2), 9(1). Anal Segment: Saddle moderately pigmented ex-
cept for darker basal ring; covered with small spicules; caudal margin with serrated
spines of varying length. Gills long, subequal, as long as ventral brush. Hair 1(2), 2(6,
6-7), 3(1), 4a(5, 4-5), 4b(3, 3-4), 4c(3, 3-4), 4d(4, 3-4), 4e(3, 3-4), 4f(3, 1-3).

SYSTEMATICS. A. spinosus is recognized as a distinct species entirely on the
basis of striking and constant features of the immature stages as indicated in the diag-
nosis. The single individually reared adult (?) shows no obvious differences from the
5 individually reared specimens of brevis. Therefore, the identity of the male and the
5 females listed below is questionable; they could possibly be brevis. A. spinosus is
known at present only from areas of cloud forest above 4,000 feet in Costa Rica and
Panama. It is sympatric throughout this range with the dominant quadrivittatus. Al-
though the 2 species are reported as being associated in 2 collections of immature
stages in Costa Rica (CR 13,54), it is not known if they were actually found together
in the same bromeliads. In Panama, spinosus has been collected in the same locality
as brevis but in this instance the immature stages were not associated in a given col-
lection, suggesting an ecological barrier between the 2 species.

BIONOMICS. Immature stages of spinosus occur in epiphytic bromeliads. Two fe-
males from Panama provisionally identified as spinosus were taken in biting-landing
collections.

DISTRIBUTION (fig.5). Costa Rica and Panama (Chiriqui mountains) at eleva-
tions above 4000 ft. Material examined: 85 specimens; | 6, 6 2, 3 pupae, 75 larvae;
2 individual rearings (1 larval, 1 incomplete).

COSTA RICA. Alajuela: Volcan Poas, 14 Apr 1938, 2 1 [USNM]. Cartago: Cervantes, 9 Nov
1962, J.N. Belkin, C.L. Hogue and W.A. Powder (CR 13, type series), 1 Ip? (13-111), 4 L [USNM,
UCLA]. Cipreses, 1 9 [USNM]. La Sierra, elev. 6300 ft, 24 Nov 1962, C.L. Hogue and W.A. Pow-
der, 55 L(CR 54) [UCLA]. Locality not specified: C. Picado, 1 6,2 9 [UCLA, USNM].

PANAMA. Bocas del Toro: La Sierra, elev. 5400 ft, 8 Apr 1963, A. Quinonez, 5 L (PA 180),
1 9 (PA 181) [UCLA]. La Zorra, elev. 4400 ft, 5 Apr 1963, A. Quinonez, 1 L (PA 172); 6 Apr

1963, A. Quinonez, 1 IP (176-103), 1 P, 1 L (PA 176); 7 Apr 1963, A. Quinonez,5 L(PA 179); _ .

9 Apr 1963, A. Quinonez, 1 9 (PA.184) [UCLA].

Sexlineatus Group

FEMALE. Head: Median light line on vertex with narrow scales. Proboscis dis-
tinctly longer than forefemur. Torus with restricted mesal scale patch. Thorax: Meso-
notal light lines mostly white, rarely creamy; inner and outer dorsocentral lines com-
plete, subparallel; lateral marginal line continuous; scales on anterior promontory
white. Antealar area above paratergite with dark scales. Paratergite with a moderate
scale patch extending entire sclerite length. Upper ppn entirely covered with scales,

Berlin: Subgenus Howardina 87

both dark and light scales present; upper stp patch quadrate. Legs: Distal white spot
on anterior surface of midfemur and hindfemur present. Tibiae predominantly dark,
with light scales in narrow longitudinal lines. Wing: Veins entirely dark scaled. Abdo-
men: Basomedian tergal light markings present on III-VII.

MALE. Palpus with basomedian light scales on segments 4,5.

MALE GENITALIA. Lobe of tergite IX prominent, with 5,6 strong bristles. Side-
pieces separated at base by a distance less than width of aedeagus; without any indi-
cation of lobes on tergomesal or sternomesal margins. Claspette represented by a
conical tubercle. Aedeagus shorter than proctiger; apex smoothly rounded, without
any indication of lateral lobes.

PUPA. Cephalothorax: Trumpet moderate, index about 5.0, distinctly flared api-
cally; hairs 10,12-C usually with a few branches. Abdomen: Hair 5-IV-V moderate,
not exceeding the apex of second tergite following; hair 9-VII moderate to strong,
distinctly larger than hair 6, exceeding lateral angle of tergite. Paddle: Short; apex
rounded. ,

LARVAE. Head: Hair 4-C weak, lightly pigmented, at least 4-branched; 5-C sin-
gle, moderate, strongly pigmented; 6-C heavy, darkly pigmented, triple or double; 7-
C moderate, double or triple; 15-C short, barely exceeding base of mental plate. Tho-
rax: Roundish in outline; integument strongly spiculate, spicules hairlike, simple.
Abdomen: Integument densely spiculate. Hair 8-VII slightly cephalodorsad of 9-VII.
Siphon: Integument imbricate; acus distinct, attached. Anal Segment: Hair 3-X dou-
ble; ventral brush with 5 pairs of hairs.

DISCUSSION. Only 1 species is recognized at this time in the Sexlineatus Group
although it is evident, as indicated under the species, that there is considerable dif-
ferentiation in some populations.

The Sexlineatus Group shows similarities with the Whitmorei Group in adult orna-
mentation (dark-scaled antealar area) and with the Allotecnon Group in the larva
(head hair 15-C short). It is the most widespread of all the groups in the section and
occurs at lower elevations than any other, particularly in the southern part of its
range in Trinidad and in lowland Ecuador where it seems to be the only species of
Howardina present (fig.4). In the central (Honduras, Panama) and northern (Mexico)
portions of its range it occurs at moderate to high elevations and may be sympatric
with the Allotecnon and Quadrivittatus Groups.

21. Aedes (Howardina) sexlineatus (Theobald)
Figs.4,9 43,44

1901.Stegomyia sexlineata Theobald, 1901:308. TYPE: Holotype ? (102), Agua Santa (St. George),
Trinidad, F.W. Urich [BM].

1907.Stegomyia dominicii Rangel and Romero-Sierra, 1907:48. TYPE: ?, Hacienda El Paraiso on
Rio Anauco [Caracas], Venezuela [LU]. NEW SYNONYMY.

Aedes (Howardina) sexlineatus of Dyar (1928:229); Edwards (1932:155); Martini (1935:49); Komp
(1936a:57); Lane (1939:101); Martinez and Prosen (1950:38-39); Lane (1953:715-716); Hors-
fall (1955:471); Martinez and Prosen (1955:23-25); Reyes (1955:74,89); Trapido, Galindo and
Carpenter (1955:530,534,537); Vargas (1956:22); Galindo and Trapido (1957:122,124,125,
134); Trapido and Galindo (1957:134); Belkin, Schick and Heinemann (1965:69); Forattini
(1965:113,232); Cova Garcia, Sutil and Rausseo (1966a:58,287; 1966b:74,141 334-335).

Aedes (Howardina) sexlineata of Bonne and Bonne-Wepster (1925:367-368).

88 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Aedes sexlineatus of Urich (1913:527,528); Howard, Dyar and Knab (1917:847-849); Dyar (1918:
73); Cerqueira (1943b:33); Boshell-Manrique and Osorno-Mesa (1944:173,174,177,178); Del
Ponte, Castro and Garcia (1951:240); Aitken (1960:4).

Aedes (Soperia} sexlineata of Anduze (1941:14).

Stegomyia sexlineata of Giles (1902:377); Theobald (1903b:238,239); Blanchard (1905:260).

Gymnometopa sexlineata of Coquillett (1905:183; 1906b:25); Theobald (1907:210; 1910:219);
Surcouf and Gonzalez-Rincones (1911:212-213).

Haemagogus sexlineata of Dyar and Knab (1906a:166).

Aedes (Howardina) dominicii of Bonne and Bonne-Wepster (1925:376); Dyar (1928:231); Ed-
wards (1932:155).

Aedes (Soperia) dominicii of Belkin, Schick and Heinemann (1965:74); Cova Garcia, Sutil and
Rausseo (1966b:80).

Stegomyia dominicii of Surcouf and Gonzalez-Rincones (1911:131,134-135; 1912:286-287).

FEMALE. Wing: 2.8 mm. Proboscis: 2.2 mm. Forefemur: 1.9 mm. Abdomen:
about 2.9 mm. Head: Light decumbent scales white. Erect scales whitish on occiput,
dark brown on sides and disc. Palpus about 0.2 of proboscis, white scales on mesal
surface and apex of segment 4. Torus light brown laterally, darker mesally and with
white scales. Thorax (fig.9): Mesonotal light lines white to dingy white; inner and
outer dorsocentral lines complete from anterior margin to scutellum, scales at level
of prescutellar area lighter; lateral marginal line short, narrow from scutal angle to
wingbase, scales longer, broader at wingbase. Antealar area above paratergite com-
pletely dark scaled. Median scutellar lobe with moderately broad dark scales laterally
and with median narrow white scales from base to apex; lateral lobes with narrow
dark scales. Paratergite scales broad, white. Apn scales white, predominantly broad,
a few anterior scales narrower; upper ppn with narrow black scales on upper half,
light to creamy narrow and broad scales on lower half, extent of black scales varied
in different populations; psp scales distinct, narrow and creamy in front, broad and
white behind; other scale patches pure white, distributed as in group. Legs: Forefe-
mur dark anteriorly except for basal creamy scales, dark posteriorly except for a nar-
row creamy longitudinal line; midfemur mostly dark anteriorly except for basal
creamy scales, a narrow mesal white line to middle and a white spot at distal 0.67,
proximal 0.6 of posterior surface white, rest dark scaled; hindfemur predominantly
dark anteriorly except for broad white line to 0.75 terminating in a white spot, pos-
terior surface white to 0.75, rest dark scaled. Foretibia and midtibia predominantly
dark except for anterior, narrow, faint white line; hindtibia dark. Foretarsus with
white scales on segments 1,2; midtarsus with short incomplete white rings on 1,2,
often a few white scales on 3; hindtarsus with broader white rings on 1-3, occasion-
ally some white scales on segment 4 and sometimes on 5. Haltere: Stem pale, knob
dark basally, fringe scales white. Abdomen: Basomedian tergal scales creamy, silvery
on VII. Sternites with broad basal V-shaped pure white patches; apical white scales
apparent up to segment V.

MALE. Wing: 2.61 mm. Proboscis: 2.43 mm. Forefemur: 1.8 mm. eRe sim-
ilar to female in color.

MALE GENITALIA (fig.43). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with 5,6 moderate setae. Sidepiece length about 3.0 of medi-
an width. Clasper about 0.5 of sidepiece, spiculate in basal 0.33. Aedeagus short,
apex uniformly sclerotized, smoothly rounded, without indication of apical lateral
lobes.

PUPA (fig.43). Abdomen: 3.6 mm. Trumpet: 0.43 mm; index about 5.0. Paddle:
0.6 mm. Diagnostic characters as in the key. Chaetotaxy as given below, hairs moder-

Berlin: Subgenus Howardina 89

ately to darkly pigmented. Cephalothorax: Mesonotum moderately pigmented be-
tween trumpets, lateral areas and metanotum slightly darker; trumpets contrasting.
Hair 1(1), 2(1), 3(1), 4(3, 2-5), 5(2, 2-3), 6(1, 1-2), 7(2, 2-3), 8(2, 1-4), 9(1), 10(4,
1-7), 111, 1-2), 12(3, 1-4). Abdomen: Tergite I uniformly strongly pigmented ex-
cept for usual light membranous area; segments II-VIII with imbricate integument,
strongly pigmented, progressively lighter caudad. Segment I: hair 1(6-13 primary
branches, each with many secondary ones), 2(1, 1-2), 3(1, 1-2), 4(2, 2-4), 5(3, 1-6),
6(1, 1-2), 71, 1-4), 9(1, 1-3), 10(1-2; present in some), 11(1-4, present in some).
Segment IT: hair 0(1), 1(2, 1-7), 2(1, 1-2), 31, 1-2), 4(2, 1-4), 5(1, 1-2), 6C1, 1-2),
7(1, 1-2) present in some, 9(1), 10(1, present in some), 11(1-2, present in some). Seg-
ment III: hair 0(1), 1(3, 1-5), 21, 1-3), 3d, 1-2), 4(2, 1-3), 5(1, 1-2), 6(1), 7(2, 1-3),
8(3, 2-4), 9(1), 101, 1-3), 11(1), 14(1). Segment IV: hair 0(1), 1(2, 2-5), 2(1), 3(2,
2-4), 4(2, 1-2), 5(2), 6(1), 7(2, 1-2), 8(2, 1-3), 9(1), 10(1), 11(1), 14(1). Segment V:
hair 0(1),°1¢2,.1-3), 2@h, 1-2);:301; 1-3)40; 1-3); 502), 601), 102 1-4); 8, 13),
9(1), 10(1), 11(1), 14(1). Segment VI: hair 0(1), 1(1, 1-2), 2(1), 3(1), 41, 1-2), 5,
1-2), 6(1), 7C1, 1-2), 8(2, 1-2), 9(1), 101, 1-2), 11(1), 14(1). Segment VII: hair 0(1),
I(1, 1-2), 2(1), 31), 4€1), 51, 1-3), 6(2, 1-3), 7(1), 8(3, 1-4), 9G, 2-5), 10(1), 11(1),
14(1). Segment VIII: hair 0(1), 4(1), 9(6, 4-9), 14(1). Paddle: Uniformly moderately
pigmented, midrib darker; external margin serrated, apex smooth; hair 1-X slender.
Male genital lobe extending to 0.59 of paddle, female genital lobe to 0.41.

LARVA (fig.44). Head: 0.76 mm. Siphon: 0.65 mm. Anal Saddle: 0.27 mm. Chae-
totaxy as given below, hairs moderately to darkly pigmented. Head: Mental plate with
a strong median tooth and 9-11 distinct teeth on either side, often a smaller detached
tooth at base. Hair 0(1), 1(1), 3(1), 4(7, 4-8), 5(1, strong), 6(3, 2-3), 7(3, 2-3), 8(1),
9(2, 2-3), 10(2, 1-2), 11(17, 10-19, stellate), 12(1), 13(1), 14(12, 12-24, stellate),
15(2, 2-4). Antenna: Length about 0.3 of head; pigmentation moderate; shaft with a
few scattered denticles. All hairs except 1-A single; hair 1-A (2, 2-3) about 0.6 from
base. Thorax: Integument strongly pilose, spicules simple, hairlike; apices of long
hairs attenuate and weakly barbed. Stellate hairs with many barbed branches. Pro-
thorax: hair 0(23, 19-25), 1(5, 2-8), 2(1, 1-2), 3(20, 18-26), 4(1, 1-3), 5(1), 6(1), 7(2,
2-3), 8(20, 15-23), 9(17, 16-24), 10(1), 11(1, 1-2), 12(1), 13(20, 18-29), 14(32, 30-
42). Mesothorax: hair 1(11, 8-15), 2(1), 3(1), 4(1), 5(1), 6(4, 3-5), 7(1), 8(5, 4-6),
9(5, 4-6), 10(1), 11(2, 1-4), 121, 1-2), 13(17, 12-21), 14(16, 10-20). Metathorax:
hair 1(21, 9-21), 2(1), 3(8, 3-9), 4(13, 12-18), 5(11, 9-16), 6(1), 7(4, 3-5), 8(17, 12-
19), 9(3, 2-5), 10(1), 11(2, 2-4). 12(1), 13(22, 16-26). Abdomen: Integument and
stellate hairs as on thorax. Spicules as on thorax, but longer and more profuse on
ventral side of segments. Segment I: hair 1(20, 12-21), 2(20, 18-25), 3(1), 4(1), 5(15,
13-20), 6(2, 2-3), 7(1), 9(7, 5-10), 10(1), 11(20, 12-23), 13(15, 9-16). Segment II:
hair O(1), 1(20, 16-26), 2(20, 16-24), 3(1), 4(1, 1-2), 5(16, 12-17), 6(2, 2-3), 7(9, 5-
11), 8(1), 9(10, 6-13), 10(1, 1-2), 11€1, 1-2), 12(1), 13(14, 12-18). Segment III: hair
O(1), 1(20, 16-23), 2(20, 9-24), 3(1), 4(1), 5(14, 14-20), 6(2), 7(14, 11-17), 8(1),
9(14, 10-18), 10(1), 11€1), 12(1), 13(13, 11-17), 14(1). Segment IV: hair 0(1), 1(20,
12-21), 2(22, 13-25), 3(1), 4(1), 5(15, 12-20), 6(2), 7(13, 10-15), 8(1), 9(18, 11-19),
101), 1101), 121), 13C11, 9-17), 141). Segment V: hair 0(1), 1(17, 12-21), 2018.
17-24), 3(1), 4(1), 5(15, 13-20), 6(2), 7(12, 10-15), 8(1), 9(16, 13-19), 10(1); 11(1),
12(1), 13(10, 9-14), 14(1). Segment VI: hair 0(1), 1(17, 12-20), 2(18, 17-22), 3(1),
4(1), 5€16, 13-17), 6(1, 1-2), 7(11, 8-12), 8C1), 9(17, 11-19), 10(2), LTC y, 12).
13(12, 8-15), 14(1). Segment VII: hair 0(1), 1(18, 12-18), 2(18, 15-21), 3(1), 4(1),
5(13, 11-17), 6(8, 8-12), 7(1), 8, 8-10), 9(12, 12-21), 10(1), 11(1), 121), 13(9, 7-
12), 14(1). Segment VII: Comb scales in a single row, usually 10(8-14) in number;

90 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

individual scale long, pointed apically and fringed nearly to tip. Hair 0(1), 1(20, 17-
24), 2(1), 3(7, 5-8), 4(1), 5(15, 12-17), 14(1). Siphon: Index about 2.1-2.4; integu-
ment moderately pigmented, imbricate. Pecten extending to 0.42-0.45, with 13-20
teeth; individual tooth long, pointed, with basal denticles. Hair 1(2, 2- 3): 2(1), 3-5
not visible, 6(1), 7(1), 8(1, 1-2), 9(1). Anal Segment: Saddle moderately pigmented
and weakly spiculate; caudal margin with spines of varying length. Gills moderate,
about 1.5-2.0 of saddle length, ventral slightly shorter than dorsal. Hair 1(4, 3-5),
2(5, 3-6), 3(2), 4a(3, 2-4), 4b(2, 2-3), 4c(2, 2-3), 4d@2, 2-3), 4e(2, 2-3).

EGG. Length about 0.73 mm, width 0.1 mm; exochorion transparent, with cir-
cular clear refractile bodies; endochorion dark, with irregular polygonal sculpturing.

SYSTEMATICS. Formerly the name sexlineatus was applied to quite unrelated
forms, some of which are now recognized as distinct species in other groups of the
Sexlineatus Section. In the restricted sense used here, with the diagnostic features
mentioned under the group and the species, sexlineatus appears to be a complex of
several closely related forms. It is impossible at this time to resolve the complex
owing to the paucity or total absence of associated immature stages and even adults
from critical areas. The best diagnostic features of the complex are: (1) in the adults,
the presence of dark scales on the upper ppn and antealar area; and (2) in the larvae,
the presence of a dense vestiture of hairlike spicules on the thorax and abdomen; the
aedeagus of the male genitalia may also prove to be distinctive. I have seen samples
from populations with these characters from central and southern Mexico, Honduras,
Panama, Ecuador, Venezuela and Trinidad. The species has been reported from inter-
mediate areas but little reliance can be placed on these records although probably the
complex does occur throughout Central America and in Colombia.

The description of the species is based on topotypic associated material from Trin-
idad. The females and 4 males from Venezuela are so similar to those from Trinidad
except for slightly yellowish mesonotal lines that there is no question that they be-
long to the same taxon. No significant departure was noted in the limited material of
immature stages from Venezuela. From the original description and illustration it is
evident that dominicii Rangel and Romero, 1907 from the Rio Anauco is based on
this population and it is therefore synonymized here with sexlineatus. The Trinidad
and Venezuela populations appear to be confined to the coastal areas and are known
from near sea level to moderate elevations in the coastal ranges.

The Ecuador population is known only by males and females from the lowlands
of the Guayas River basin. These specimens resemble the typical populations except
for fewer dark scales on the upper ppn and a tendency toward more extensive light
scaling on the tarsi, some specimens showing some white scales on segment 3 of the
foretarsus, segments 3 and/or 4 of the midtarsus and segments 4 and/or 5 of the
hindtarsus. In the male genitalia, some of the setae of the ninth tergite lobe are slight-
ly flattened, approaching the condition characteristic of ecuadoriensis of the Elea-
norae Group.

Limited material of all stages has been examined from Panama. While the majority
of adults agree with the typical populations, 2 show a preponderance of white scales
on the upper ppn. I have not been able to detect any significant differences in the
immature stages from the typical populations.

The 3 females from Honduras, 2 females from Colombia and all adults from Mexi-
co differ from the typical populations apparently only in the presence of fewer dark
scales cn the upper ppn. In the limited material of immature stages from Mexico no
significant differences from typical populations were noted. These populations are
found at higher elevations than the typical ones, particularly the ones from Cuerna-

Berlin: Subgenus Howardina 91

vaca, Mexico (about 7,000 feet).

In the southern portion of the range, populations of sexlineatus are apparently
largely or entirely restricted to the rain forest and tropical deciduous forest at low
and moderate elevations. This may be the case even in Central America and Mexico
where sexlineatus populations are known only from higher elevations where patches
of such forests may be present.

BIONOMICS. Immature stages of sexlineatus have been collected only in epiphy-
tic bromeliads. Females are readily attracted to human bait and Trapido and Galindo
(1957) found that in Panama they bite both at ground level and in the forest canopy.
Attempts at experimental infection with the yellow fever virus have yielded negative
results in Trinidad (Aitken 1960:4). Similar negative results have been reported by
Boshell-Manrique and Osorno-Mesa (1944:174,177,178) but it is doubtful that the
Species involved was actually sexlineatus.

DISTRIBUTION (fig.4). Mexico (southern part), probably Guatemala, Honduras,
Nicaragua, possibly Costa Rica, Panama, Colombia (upper Orinoco drainage), Vene-
zuela (northern), Trinidad and Ecuador (Guayas River basin) at elevations of 250-
7006 ft. Material examined: 1099 specimens; 28 6, 684 ?, 35 pupae, 352 larvae; 32
individual rearings (1 pupal, 24 larval, 7 incomplete).

COLOMBIA. Boyaca: Cuincha, 2 9; Guamales, 2 9; San Isidro, 3 9 (Boshell-Manrique and Osor-
no-Mesa 1944:174,177,178). Meta: Retiro, 15-16-19 Aug 1935, 2 9 [UCLA, USNM].

_ ECUADOR. Babahoyo, Aug 1938, F. Campos, 4 6, 18 2 [USNM] ; Balzapamba, Aug 1938, F.
Campos, 4 d, 1 2 [USNM], 6 ? (KO 209A-21), 1939 or 1941, H. Hansen, 2 ? (KO 209A-14)
[UCLA] . Guayaquil, J. Murdock, 2 6,3 9 [USNM].

HONDURAS. Tegucigalpa, 15 July 1945, 3 9 [USNM]. :

MEXICO. Chiapas: San Cristobal (Martini 1935:49). Tuxtla, 23 July 1963, E. Fisher (MF 8),
21 2; Sumidero, 17 Aug 1964, D. Verity, 1 lpd (117-11), 4 L (MEX 117), 1 lpd (118-10); 17 Aug
1964, E. Fisher and D. Verity, 1 9? (MEX 120), 1 9 (MEX 121), 8 9 (MEX 128) [UCLA]. Morelos:
Cuernavaca, elev. 7000 ft, 29 July 1961, R. and K. Driesbach, 2 6, 2 9 [USNM]. Veracruz: Cordo-
ba, 12 July 1964, E. Fisher and D. Verity, 5 9 (MEX 18), 1 9? (MEX 19), 1 9 (MEX 21); 13 July
1964, E. Fisher and D. Verity, 3 9 (MEX 26), 1 Ip¢ (28-14), 1 Ip? (28-11); 16 July 1964, E. Fisher
and D. Verity, 1 L (MEX 38) [UCLA].

NICARAGUA. Villa Somoza, Aug, Sept, Nov 1953, 1 9 (Galindo and Trapido 1957:146).

PANAMA. Bocas del Toro: Almirante, 1952-53, 77 A (Trapido and Galindo 1957:122,125).
Cocle: El Valle, elev. 2000 ft, 14 Aug 1963, A. Quinonez (PA 533), 1 IP (533-101); Valle de Anton,
elev. 2400 ft, 13 Aug 1963, A. Quinonez (PA 504), 3 Ip (504-101,102,108), 1 Ip? (504-104), 1 pd
(504-103), 1 IP (504-106) [UCLA]. Tucue, elev. 1000 ft, 12 June-14 Dec 1950, 2 A (Trapido,
Galindo and Carpenter 1955:530). Panama: Campana, elev. 2000 ft, 14 June-6 Dec 1950, 24 A
(Trapido, Galindo and Carpenter 1955:530).

TRINIDAD. Caroni: Brasso Venado, elev. 400 ft, 21 June 1964, R.L. Manuel and R. Martinez,
24 9 (TR 513) [UCLA]. Mayaro: Bush Bush Forest, 9 Dec 1963, TRVL, 2 9 (TR 62); 20 Dec
1963, TRVL, 1 ? (TR 5); 27 Dec 1963, TRVL, 6 2? (TR 21); 30 Dec 1963, TRVL, 2 2 (TR 10);
17 Jan 1964, TRVL, 4 9° (TR 28); 22 Jan 1964, TRVL, 1 2 (TR 40) [UCLA]. Nariva: Archers
Estate, 5 Nov 1964, A. Guerra, 1 9? (TR 819) [UCLA]. Biche, elev, 250 ft, 9 July 1964, F. Powd-
har, 3 9 (TR 563) [UCLA]. Charuma Forest, elev. 500 ft, 27 Aug 1964, A. Guerra, 24 9 (TR 644);
8 Oct 1964, A. Guerra, 3 2? (TR 761) [UCLA]. St. Andrew: Caratal Road, elev. 250 ft, 3 Dec
1964, F. Powdhar, 1 ? (TR 866) [UCLA]. Coryal, elev. 250 ft, 18 June 1964, A. Guerra, 29(TR
498) [UCLA]. Mt. Harris, 23-31 July 1924, C.L. Withycombe, 1 9 [USNM]; elev. 250 ft, 16 July
1964, F. Powdhar, 16 9? (TR 573) [UCLA] . Nestor Village, elev. 100 ft, 12 June 1965, A. Guerra,
1 2 (TR 484) [UCLA]. Tamana, 1945, 1 9 [USNM];Mt. Tamana, elev. 250 ft, 19 June 1964, A.
Guerra, 43 ? (TR 512); 20 Aug 1964, A. Guerra, 1 9? (TR 619); elev. 1000 ft, 23 Aug 1964, R.L.
Manuel and R. Martinez, 5 2? (TR 620) [UCLA]. Valencia, Cumaca, elev. 1000 ft, 3 Sept 1964, A.
Guerra, 2 9 (TR 654); El Quemado Road, elev. 250 ft, 3 July 1964, F. Powdhar, 1 9 (TR 547)
[UCLA] . St. David: Sans Souci, elev. 250 ft, 5 Mar 1964, A. Guerra, 1 9 (TR 151) [UCLA]. St.

92 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

George: Aripo Valley, elev. 500 ft, 16 Apr 1964, A. Guerra, 1 2 (TR 330) [UCLA] . Blanchisseuse,
2 Aug 1945, 18 9; 2 Sept 1945, 1 9 [USNM]; 1 9 (KO 209A-36) [UCLA] ; road to Blanchisseuse,
2 Mar 1945, 2 9 [USNM]. Guanapo Valley, elev. 1500 ft, 17 Apr 1964, A. Guerra, 1 ? (TR 339)
[UCLA]. Heights of Guanapo, elev. 800 ft, 26 Mar 1964, A. Guerra, 1 9 (TR 259); elev. 1000 ft,
22 Apr 1965, A. Guerra, 63 2 (TR 1129) [UCLA]. La Laja Road, elev. 1000 ft, 11 Mar 1965, A.
Guerra, 44 ? (TR 1036) [UCLA]. La Laja Trace, 11 Mar 1965, 2 Ipd (1041-22,41), 1 Ip? (1041-
18), 2 lp (1041-21,23), 2 P, 33 L, 13 1 [UCLA]. Las Lapas Trace, elev. 2000 ft, 3 Apr 1964, A.
Guerra, 12 9 (TR 281) [UCLA]. Mt. Beck, elev. 1000 ft, 29 Apr 1965, A. Guerra, 1 9 (TR 1132)
[UCLA]. Port of Spain, 26 Aug 1945, 1 @ [USNM]. Simla, 25 Mar 1965, M.J. Moody, 6 ° (TRM
15) [UCLA]. Springhill Estate, elev. 1200 ft, 25 Mar 1965, A. Guerra and M.J. Moody (TRM 12),
2 Ip? (12-50,52), 10 L; 25 Mar 1965, A. Guerra, 2 lp? (1056-11,14), 1 P,5 L,11(TR 1056), 3 Ip?
(1063-31,81,82), 6 2, 86 L(TR 1063), 2 Ipd (1064-11,12), 1 9, 3 L(TR 1064) [UCLA]. Verdant
Vale, elev. 1000 ft, 10 Sept 1964, A. Guerra, 5 9? (TR 686); elev. 500 ft, 12 Nov 1964, A. Guerra,
21 2(TR 831); 9 Apr 1965, elev. 1200 ft, A. Guerra, 3 ?, 132 L(TR 1098), 1 Ipd (1099-33), 3 Ip?
(1099-11 21,22), 3 IP (1099-12,31,32), 34 L [UCLA]. County not specified: June 1905, A. Busck,
15 9; A. Busck, 7 9; F.W. Urich, 1 9 [USNM]. :

VENEZUELA. Anzoategui: Puerto La Cruz, Sept 1944, W.H.W. Komp, 1 ? (KO 209A-15),
2 6 (KO 209A-20); 26 Sept 1944, W.H.W. Komp, 10 ? (KO 209A-24), 23 2 (KO 209A-27), 24A
(KO 209A-35) [UCLA]. Aragua: Guamitas, elev. 1200 m, 27 Jan 1927, M.N. Tovar, 3 2; 24 Aug
1927, M.N. Tovar, 136 9; 27 Aug 1927, M.N. Tovar, 1 ? [USNM]. La Providencia (Cova Garcia,
Sutil and Rausseo 1966b:74). Maracay, 2 July 1927, M.N. Tovar, 2 9;M.N. Tovar, 1 9 [USNM].
Rancho Grande, 2 Nov 1962, I. Ortiz C., 7 9? [USNM]. Turmero (Cova Garcia, Sutil and Rausseo
1966b:74). Distrito Federal: Caracas, 21 July 1938, 1 6;P.J. Anduze, 1 6, 1 9 [USNM]; Hacienda
El Paraiso, 9 (Rangel and Romero-Sierra 1907). Carayaca (Cova Garcia, Sutil and Rausseo 1966b:
74). Miranda: El Hatillo (Cova Garcia, Sutil and Rausseo 1966b:74). Monagas: Guacharo, F.M.
Root, 5 2?(VZR 267) [UCLA].

Whitmorei Group

FEMALES. Head: Median light line on vertex with moderately broad scales. Pro-
boscis subequal to forefemur in length. Torus with an almost complete ring of white
scales broken only dorsally. Thorax: Mesonotal light lines usually white to silvery;
inner dorsocentral line complete; outer dorsocentral line shorter, not reaching ante-
rior margin, diverging outwards anteriorly; lateral marginal line broken into narrow
prescutal, antealar and supraalar patches; scales on anterior promontory white. An-
tealar area above paratergite with dark scales. Paratergite and ppn with restricted
light patch confined to posterior half; upper stp patch narrow, elongate. Legs: Distal
white spot on anterior surface of midfemur present, absent on hindfemur. Tibiae
predominantly dark, with light lines or scattered scales. Wing: Veins entirely dark
scaled. Abdomen: Basomedian tergal light markings absent.

MALES. Palpus predominantly dark, white scales restricted to base of segments 4
and 5.

MALE GENITALIA. Lobe of tergite [IX prominent, with a few (5-7) strong or
many (about 12) weak bristles. Sidepieces separated at base by a distance greater
than width of aedeagus; tergomesal and/or sternomesal margins with more or less
distinct median lobes. Claspette completely absent. Aedeagus moderate to long,
shorter than or subequal to proctiger; apex deeply emarginate, with distinct apical
horns.

PUPA (whitmorei only). Cephalothorax: Trumpets short, index about 3.0, not
expanded apically; hair 10-C branched, 12-C single or double. Abdomen: Hair 5-IV-V
moderate, not exceeding apex of second tergite following; hair 9-VII moderate,

Berlin: Subgenus Howardina 93

branched, exceeding lateral angle of tergite. Paddle: Moderate; apex rounded.

LARVA (whitmorei only). Head: Hairs 4-7-C moderately pigmented, all at least
5-branched; hair 15-C moderate, exceeding apex of mental plate. Thorax: Squarish
in outline; integument strongly spiculate, spicules hairlike, simple. Abdomen: Integ-
ument strongly uniformly spiculate. Hair 8-VII slightly cephalodorsad of 9-VII. Si-
phon: integument distinctly spiculate; acus distinct, attached. Anal Segment: Hair 3-
X single; ventral brush with 5 pairs of hairs.

DISCUSSION. The Whitmorei Group of 2 species is the subgenus Soperia of au-
thors, which is based on a misidentified species (see discussion under subgenus How-
ardina). This group is a very distinct one in adult ornamentation and male genitalia
but is obviously related to other groups in the Sexlineatus Section. Its closest affini-
ties seem to be with the Sexlineatus and Allotecnon Groups. The immature stages of
whitmorei are less differentiated from other groups of the Sexlineatus Section than
are those of walkeri from other groups of the Walkeri Section.

The Whitmorei Group is known at present only from Colombia, Venezuela, Bo-
livia and probably Argentina (fig.5). Although whitmorei and pseudodominicii have
been found together in Meta, Colombia, the 2 species may be largely allopatric, whit-
morei occupying the northern part of the range of the group. A. pseudodominicii
penetrates into the Amazon basin at an elevation of less than 2,000 feet in Chimore,
Bolivia and apparently into the La Plata basin at moderate elevations in northern Ar-
gentina. It is also possible that this group is represented on the eastern slope of the
Andes in Ecuador (Levi-Castillo 1953:40).

22. Aedes (Howardina) whitmorei Dunn
Figs.5 ,9,45 46

1918.Aedes whitmorei Dunn, 1918:128-130. TYPE: Lectotype 2, near emerald mines, Muzo, Co-
lombia, 1916, E.R. Whitmore [USNM; selection of Stone and Knight 1956:226].

Aedes (Howardina) whitmorei of Bonne and Bonne-Wepster (1925:375); Dyar (1928:230); Ed-
wards (1932:156); Horsfall (1955 :471).

Aedes (Soperia) whitmorei of Lane (1939:98); Anduze (1941:14; 1943:193; 1947:357); Lane
(1953:701-703); Reyes (1955:74,89); Stone and Knight (1956:226); Stone, Knight and Starcke
(1959:173); Belkin, Schick and Heinemann (1965:9); Forattini (1965 :405); Cova Garcia, Sutil
and Rausseo (1966a:57,275; 1966b:80-81,141 339).

Aedes whitmorei of Antunes (1937b:77); Senevet and Quievreux (1941:258); Boshell-Manrique
and Osorno-Mesa (1944:173,174,175,177,178).

Aedes (Soperia) dominicii of Komp (1936a:63; 1936b:73-74).

Aedes dominicii in part of Bates (1944:165 166; 1945:23; 1949:16,61,320); Bugher, Boshell-Man-
rique et al (1944:33,37).

FEMALE. Wing: 2.7 mm. Proboscis: 1.98 mm. Forefemur: 1.98 mm. Abdomen:
about 2.4 mm. Head: Light decumbent scales on vertex silvery white. Erect scales
light on occiput, darkened on sides. Palpus about 0.16 of proboscis, with silvery
white scales on apex of segment 4. Torus with silvery white scales. Thorax (fig.9):
Mesonotal light lines white; inner dorsocentral line very narrow; outer dorsocentral
line narrow but with a distinct broad silvery spot of broad scales at anterior end in
fossa; lateral marginal line broken into lateral prescutal line, a small antealar and a
larger supraalar patches, lateral prescutal nearly continuous but apparently broken at
level of humeral bristle. Median scutellar lobe laterally with broad dark scales and

94 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

with median broad silvery white scales from base to apex; lateral lobes with a few
narrow silvery white scales. Scales on paratergite broad, silvery. Pleural scales broad
and silvery; lower ppn patch adjoining upper ssp; other patches as in the group. Legs:
Forefemur predominantly dark on both sides except for narrow white longitudinal
line posteriorly; midfemur dark anteriorly except for a silvery spot at distal 0.67,
proximal 0.5 of posterior surface white, rest dark; hindfemur predominantly white
anteriorly to 0.67, distal part dark, proximal 0.5 of posterior surface white, distal
dark. Tibiae usually dark, a few whitish scales at base. Foretarsus and midtarsus with
short incomplete white rings on segments 1 ,2; hindtarsus with moderately broad ones
on 1-3. Haltere: Stem pale, knob predominantly dark, fringe scales whitish. Abdo-
men: Basomedian tergal pure white scales evident on segments V-VII. Sternites with
broad basal V-shaped pure white patches; apical light scales inconspicuous.

MALE. Wing: 2.5 mm. Proboscis: 2.34 mm. Forefemur: 1.89 mm. Essentially sim-
ilar to female in color; only apparent difference is a small ventromedian white spot
on the proboscis.

MALE GENITALIA (fig.45). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with 6,7 moderately strong bristles. Sidepiece rather short,
length about 2.0 of median width; tergomesal margin with a row of unpigmented
leaflike setae, sternomesal lobe with a dense tuft of pigmented leaflike setae on an
elevated flap. Clasper about 0.5 of sidepiece, with spicules in basomesal area; spini-
form moderately long, about 0.35 of clasper length. Aedeagus narrow, subequal to
proctiger, apex deeply emarginate, with distinct horns curved laterally.

PUPA (fig.45). Abdomen: 2.86 mm. Trumpet: 0.32 mm; index about 3.0. Paddle:
0.67 mm. Diagnostic characters as in the key. Chaetotaxy as given below, hairs mod-
erately pigmented. Cephalothorax: Mesonotum lightly pigmented between trumpets,
lateral areas slightly darker; metanotum uniformly moderately pigmented. Trumpets
short, strongly pigmented and contrasting. Hair 1(1), 2(1), 3(1), 4(3, 2-4), 5(2), 6(1),
7(2), 8(1, 1-3), 9C1, 1-2), 10(4, 2-5), 11(1), 12(2, 1-2). Abdomen: Tergite I uniform-
ly lightly pigmented, except for usual membranous areas; rest of segments imbricate
and lightly pigmented, progressively lighter caudad. Segment I: hair 1(7-12 primary
branches, each with few secondary ones), 2(1), 3(1), 4(2, 2-3), 5(1, 1-3), 6(1), 7(1),
9(1), 10(1, present in a few), 11(2, 1-4, present in a few). Segment II: hair 0(1), 1(4,
2-5), 2(1), 31), 4(@2, 2-3), 5(1, 1-2), 6(1), 7(1, 1-2), 8(1), 9(1), 10(1). Segment IIT:
hair OC] ), 1G, 2-3), 201), 3€2), 4C1, 1I-2), 5(1, 1-2), 6(1), 7@, 1-3); 8(1, 1-2), 901),
10(1), 11€1), 141). Segment IV: hair 0(1), 1(2, 2-4), 2(1), 3(2, 2-3), 4(1), 5(2), 6(1),
7(2, 2-4), 8(1, 1-2), 9(1), 101), 11(1), 14(1). Segment V: hair 0(1), 1(2, 2-3), 2(1),
3(1), 4(2, 1-4), 5(1), 6(1), 7(2, 2-5), 8(1, 1-2), 9(1), 10(1), 1101), 14(1). Segment VI:
hair O(1), 1€1, 1-3), 21), 31), 4(1, 1-2), 5(2), 6(1), 71, 1-3), 8(1, 1-2), 9(1), 10Q1,
1-2), 11(1), 141). Segment VII: hair 0(1), 1(1, 1-2), 2(1), 3(1), 4(1), 5(1), 6Q, 2-3),
7(1), 8(3, 3-4), 9(5, 4-6), 10(1), 11€1), 14€1). Segment VIII: hair 0(1), 4(1), 9(10, 8-
11), 141, 1-2). Paddle: Uniformly lightly pigmented, midrib slightly darker; external
margin serrated. Male genital lobe extending to 0.5 of paddle, female genital lobe to
0.39.

LARVA (fig.46). Head: 0.84 mm. Siphon: 0.81 mm. Anal Saddle: about 0.3 mm.
Diagnostic characters as in the key. Head: Width subequal to length; pigmentation
moderate except around mouthparts where it is deeper. Mental plate with a strong
median tooth and 11 distinct teeth on each side, often a smaller detached tooth at
base. Hair 0(1), 1(1), 3(1), 4(15, 10-16), 5(6, 5-8), 6(7, 6-7), 7(14, 11-17), 8(3, 3-6),
9(11, 8-14), 10(1), 11(20, 16-20), 12(5, 4-7), 13(1), 14(14, 11-14), 15(3, 2-3). An-
tenna: Length about 0.32 of head; shaft rather uniformly pigmented and with a few

Berlin: Subgenus Howardina 95

scattered spicules; hair 1(3, 2-3), other single. Thorax: Stellate hairs with many un-
equal, weakly barbed branches; long hairs attenuate apically and weakly barbed. Pro-
thorax: hair 0(19, 19-23), 1(5, 4-6), 2(1), 3(21, 13-23), 4(1), 5(1), 6(1), 7(2), 8(@20,
16-20), 9(14, 14-17), 10(€1), 11(€1), 12€1), 13(22, 20-25), 14(34-51). Mesothorax:
hair 1(14, 13-20), 2(1), 3(1), 41), 5(1), 6(5, 4-6), 7(1), 8(6, 6-7), 9(5, 4-6), 10(1),
11(2, 1-3), 12(1), 13(17, 10-19), 14(16, 16-21). Metathorax: hair 1(25, 20-27), 2(1),
3(2, 1-2), 418, 17-19), 5(18, 17-23), 6(1), 7(5, 4-6), 8(18, 16-22), 9(4, 3-6), 10(1),
11(2, 1-2), 12(1), 13(22, 19-24). Abdomen: Integument and stellate hairs as on tho-
rax. Segment I: hair 1(23, 20-25), 2(20, 20-25), 3(1), 4(1), 5(16, 14-23), 6(2, 2-3),
7(1), 9(8, 7-8), 10(1), 11018, 15-20), 13(22, 22-25). Segment II: hair 0(1), 1(21, 20-
24), 2(22, 18-25), 3(1), 4(1), 5(24, 18-24), 6(2, 2-3), 7(10, 9-10), 8(1), 9(14, 9-15),
101), 111), 121), 13@20, 19-24). Segment III: hair 0(1), 1(22, 19-25), 2(21, 18-
27), 3(1), 4(1), 5(19, 17-26), 6(2), 7(17, 14-20), 8(1), 9(16, 14-21), 10(1), 1101),
12(1), 13(@20, 16-26), 14(1). Segment IV: hair 0(1), 1(20, 17-24), 2(22, 20-22), 3(1),
4(1), 5(22, 18-25), 6(2), 7(15, 9-18), 8(1), 9(19, 18-23), 10(1), 11(1), 121), 1318,
14-20), 14(1). Segment V: hair 0(1), 1(22, 16-23), 2(19, 17-23), 3(1), 4(1), 5(20, 18-
23), 6(2), 7(15, 12-18), 8(1), 9(19, 18-24), 10(1), 11(1), 12(1), 1319, 16-22), 14(1).
Segment VI: hair 0(1), 1(19, 19-23), 2(19, 18-23), 3(1), 4(1), 5(21, 18-22), 6(1),
7(11, 10-18), 8(1), 9(20, 17-21), 10(1), 111), 12(1), 13(17, 13-18), 14€1, 1-2). Seg-
ment VII: hair 0(1), 1(20, 18-21), 2(18, 16-22), 3(1), 4(1), 5(16, 16-22), 6(10, 10-
13), 71), 810, 8-11), 9(16, 11-18), 10(1), 11(1), 12(1), 13(11, 10-14), 14(@2, 1-2).
Segment VIIT: Comb scales usually 12(9-12) in number, in a single row; individual
scale long, pointed apically and fringed almost to tip. Hair 0(1), 1(28, 22-28), 2(1),
3(8, 7-9), 4€1), 5(15, 10-19), 14(2, 2-3). Siphon: Index about 3.0, distinctly swollen
in middle; integument uniformly moderately to strongly pigmented, with darker
basal ring, uniformly spiculate. Pecten teeth usually 16(14-20) in number, extending
to 0.4 before hair 1-S. Hair 1(3, 2-3), 2(1), 3-5 not visible, 6(1), 7(1), 8(1, 1-2), 9(1).
Anal Segment: Saddle strongly pigmented, with darker basal ring; integument uni-
formly covered with spicules; caudal margin with spines of varying length; gills sub-
equal, long, more than 2.0 of saddle length. Hair 1(2), 2(4, 3-4), 3(1), 4a(2), 4b(2),
4c(2, 1-2), 4d(2, 1-2), 4e(1, 1-2).

SYSTEMATICS. A. whitmorei is a very strikingly differentiated species errone-
ously synonymized with dominicii by Komp (1936a:53); the latter as shown here is
undoubtedly sexlineatus (see). Its only close relative is pseudodominicii, whose im-
mature stages are unknown. In ornamentation of the adults and in male genitalia the
2 species are readily separated by the features mentioned in the diagnoses and keys.
A. whitmorei is known at present from both sides of the Cordillera Oriental in Co-
lombia and from the Cordillera de la Costa in northcentral Venezuela; it probably
occurs also in the intermediate areas in the Cordillera de Merida in western Venezue-
la. On the eastern slope of the Cordillera Oriental whitmorei is sympatric with pseu-
dodominicii whose distribution extends on down to northern Argentina along the
eastern slopes of the Andes.

BIONOMICS. Immature stages of whitmorei have been collected in epiphytic bro-
meliads at elevations of about 3,000 feet. Dunn’s (1918:130) record from ‘“‘small
heavily shaded pools of clear but apparently stagnant water” is probably erroneous.
According to Bates (1949:16,61), whitmorei (as dominicii) is a relatively common
diurnal forest mosquito in the region of Villavicencio in Colombia; the females bite
both at ground level and in tree stations; and males are often observed hovering over
men and animals in the forest. Experimental infections of this species (also as domi-
nicii) with yellow fever virus have been unsuccessful (Bugher, Boshell-Manrique et al

96 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

1944:33,37).

DISTRIBUTION (fig.5). Colombia (upper Orinoco and Magdalena drainages) and
Venezuela (Cordillera de Merida and Cordillera de la Costa) at moderate elevations.
Material examined: 104 specimens; 32 6, 43 9, 7 pupae, 22 larvae; 7 individual larval
rearings.

COLOMBIA. Boyaca: Las Animas, 500 yds from mine, 17 Aug 1916, 6 ? [USNM] . Muzo, em-
erald mines, E.R. Whitmore, 1 9; 16 Aug 1916, E.R. Whitmore, 10 , type series [USNM]. Guamales,
51 9; Tamborredondo, 1 9; Cuincha, 162 2; San Isidro, 37 2 (Boshell-Manrique and Osorno-Mesa
1944:174,175,177,178). Meta: Palucho, 3 6, 2 2? [USNM]. Restrepo, 17 July 1935, W.H.W. Komp,
1 6, 1 2 (KO 203A-10); 20 Aug 1935, W.H.W. Komp, 1 6 (KO 209A-2) [UCLA]; Aug 1935,
W.H.W. Komp, 2 6, 3 2; 9 Sept 1935, W.H.W. Komp, 1 6,2 ?, 101 [USNM], 2 6, 3 2 (KO 209A-
28) [UCLA]; 1935, W.H.W. Komp, 1 6 (KO 209A-40), i 6 (KO 209A-42), 1 d (KO 209A-44)
[UCLA]; July 1935, 2 6; Sept 1935, 2 9; 1935, 2 6; 4 June 1940, 1 6; 2 d [USNM]; Retiro, 22
Aug 1935, W.H.W. Komp, 5 6; 15-16-19 Aug 1935, 2 ? [USNM]. Villavicencio, Buena Vista, elev.
3000 ft, 9 July 1965, E. Osorno-Mesa et al (COB 68), 1 Ipd (68-30), 2 L [UCLA] ; 4 June 1942,
W.H.W. Komp, 2 6 [USNM]; Forzosa, 25 Oct 1943, M. Bates, 1 ? [USNM] ; Quebrada Colorado,
elev. 2300 ft, 9 July 1965, E. Osorno-Mesa et al (COB 67), 4 Ipd (67-20,22,23 125), 2 Ip? (67-21,
24),3 L [UCLA].

VENEZUELA. Aragua: Ocumare de la Costa, 5 July 1927, M.N. Tovar, 5 ? [USNM]. Rancho
Grande, 2 Nov 1942, I. Ortiz C., 3 9 [USNM]. Distrito Federal: Caracas (Anduze 1943:193; Cova
Garcia, Sutil and Rausseo 1966b:80). Cementerio (Cova Garcia, Sutil and Rausseo 1966b:81).
Merida: Bailadores (Cova Garcia, Sutil and Rausseo 1966b:81). Tovar (Anduze 1947:357). Tachira:
Jaurequi and Uribante (Anduze 1947:357). :

23. Aedes (Howardina) pseudodominicii Komp
Figs.5 9,58

1936.Aedes pseudodominicii Komp, 1936a:75. TYPE: Holotype 6, Restrepo, Meta, Colombia,
Aug-Sept 1935, E. Osorno-Mesa [NE].

1955.Aedes (H.) delpontei Martinez and Prosen, 1955:35-40. TYPE: Holotype 6, Chapare, Cocha-
bamba, Bolivia, elev. 6000 ft, Jan 1949, Feb 1952 or Nov 1953, Martinez and Prosen
[MEPRA]. NEW SYNONYMY.

Aedes (Soperia) pseudodominicii of Komp (1936b:63; 1956:37).

Aedes (Soperia) dominicii of Antunes (1937a:16; 1937b:76-77); Lane (1939:98); Senevet and
Quievreux (1941:258); Anduze (1943:192-193); Boshell-Manrique and Osorno-Mesa (1944:173,
174,175,177,178); Levi-Castillo (1952:555; 1953:40); Lane (1953:700-701); Reyes (1955:73,
89); Stone , Knight and Starcke (1959:173); Forattini (1965: 405).

Aedes (Soperia) dominicii in part of Bates (1944:165 166; 1945:23; 1949:16,61,320); Bugher, Bo-
shell-Manrique, et al (1944:33,37).

Aedes (Howardina) dominicii of Horsfall (1955:471).

Aedes (H.) delpontei of Martinez and Prosen (1959:603,605); Stone, Knight and Starcke (1959:
174); Prosen, Carcavallo and Martinez (1964:104,119).

FEMALE. Wing: 3.42 mm. Proboscis: 2.61 mm. Forefemur: 2.61 mm. Abdomen:
about 3.24 mm. Similar to whitmorei but usually slightly larger in size and differing
in the following. Head: Light decumbent scales white. Erect scales on occiput and
sides whitish. Palpus about 0.17 of proboscis, with a few mesal and apical white
scales on segment 4. Torus scales white. Thorax (fig.9): Mesonotal light lines white,
rarely faintly creamy; inner and outer dorsocentral line without silvery spot in fossal
area. Scutellar light scales white. Pleural scale patches white and distributed as in

Berlin: Subgenus Howardina 97

whitmorei. Legs: Forefemur predominantly auburn except for a few basal white
scales anteriorly and a narrow longitudinal line posteriorly; midfemur predominantly
auburn except for a white spot at distal 0.67 anteriorly and a mesal white line to
middle posteriorly; hindfemur predominantly white on both sides, rest mostly au-
burn; midfemur and hindfemur with a short mesal patch merging with knee spot.
Foretibia predominantly auburn, with a few scattered white scales; midtibia and
hindtibia with a continuous white line, rest auburn. Foretarsus with a few white
scales on segment 1; midtarsus with short incomplete white ring on 1; hindtarsus with
broader white rings on segments 1,2. Haltere: Stem brownish, knob predominantly
auburn, with a few white fringe scales. Abdomen: Basal sternal light scales white,
apical white scales apparent to segment V, rest auburn.

MALE. Wing: 2.52 mm. Proboscis: 2.34 mm. Forefemur: 1.9 mm. Essentially sim-
ilar to female in color. Proboscis completely dark.

MALE GENITALIA (fig.58). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with many (about 12) narrow bristles. Sidepiece rather short,
broader at the middle, tergomesal area with few and sternomesal area with numerous
narrow setae. Clasper about 0.6 of sidepiece, stout, curved; spiniform short. Aedea-
gus shorter than proctiger; dorsal sclerotization deeply emarginate, apical horns dis-
tinct, curved upwards and dorsally.

PUPA, LARVA unknown.

SYSTEMATICS. A. pseudodominicii is a distinct, clearly marked species errone-
ously synonymized by Antunes (1937a:16) with dominicii; the latter is actually sex-
lineatus (see). On the basis of ornamentation of the adults (only stages known), pseu-
dodominicii is evidently closely related to whitmorei, which has also been confused
(Komp 1936a:53; 1936b:73-74) with dominicii. The 2 species are easily distinguished
by the diagnostic features given above. A. pseudodominicii apparently occurs at mod-
erate elevations only along the eastern slopes of the Andes, from the Cordillera Ori-
ental in Colombia, where it is sympatric with whitmorei, all the way down to north-
ern Argentina. It penetrates the upper reaches of the 3 main eastern drainage sys-
tems of South America, the Orinoco, Amazon and La Plata. Although I have not seen
any of the type material of delpontei Martinez and Prosen, 1955, it is evident that
this population from Bolivia is conspecific with the typical population from Colom-
bia as the original description of delpontei obviously matches in all pertinent details
the diagnostic features of pseudodominicii as do 2 females collected by Torres-Munez
at the type locality of delpontei. I have not seen the female paratype of delpontei
from Argentina but see no reason to doubt its conspecificity with the material from
Bolivia and Colombia.

BIONOMICS. The immature stages of pseudodominicii are unknown but undoubt-
edly will be found in epiphytic and/or terrestrial bromeliads. It is very likely that the
reports of Bates (1949) and Bugher and Boshell-Manrique et al (1944) mentioned un-
der whitmorei apply at least in part to pseudodominicii since at that time both spe-
cies were probably identified as dominicii.

DISTRIBUTION (fig.5). Colombia (upper Orinoco and Magdalena drainages), Bo-
livia (upper Amazon drainage) and Argentina (upper La Plata drainage) at moderate
elevations. Material examined: 20 specimens; 10 6, 10 2; no individual rearings.

ARGENTINA. Salta: Campamento San Pedro S.0.C., Departamento San Martin, 1 ? paratype
of delpontei (Martinez and Prosen 1955:40).

BOLIVIA. Cochabamba: Chapare, Chimore, 29 June 1944, Torres-Munoz, 1 ?; 29 Apr 1944,
1 9 [USNM]. Za Paz: Nor Yungas, A (Prosen, Carcavallo and Martinez 1964:104).

COLOMBIA. Boyaca: La Bodega, 22 ?;Guamales, 58 2; Tamborredondo, 33 2; Cuincha, 19 9;

98 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

San Isidro, 15 ? (Boshell-Manrique and Osorno-Mesa 1944:173,174,175,177,178). Meta: Palucho,
2 6 [USNM]. Restrepo, 26 July 1935, W.H.W. Komp, 1 ? (KO 209A-39); 20 Aug 1935, W.H.W.
Komp, 4 6 (KO 209A-2) [UCLA]; 7 Sept 1935, E. Osorno-Mesa, 1 6; 1935, W.H.W. Komp, 1 d;
July 1935, 1 9 [USNM]; Retiro, 26 July 1935, W.H.W. Komp, 3 ? (KO 209A-31) [UCLA] ; 15-19
Aug 1935, 16, 2 9 [USNM]. Villavicencio, Buena Vista, 4 June 1942, W.H.W. Komp, 1 d [USNM].

Allotecnon Group

FEMALES. Head: Median light line on vertex with narrow scales. Proboscis dis-
tinctly longer than forefemur. Torus with a restricted mesal scale patch. Thorax:
Mesonotal light lines golden; inner and outer dorsocentral lines complete, subparallel;
lateral marginal line continuous; scales on anterior promontory creamy to golden.
Antealar area above paratergite with golden scales. Paratergite with a moderate scale
patch extending entire sclerite length. Upper ppn entirely covered with light scales;
upper stp patch narrow, elongate. Legs: Distal white spot on anterior surface of mid-
femur and hindfemur present. Tibiae completely dark. Wing: Veins entirely dark
scaled. Abdomen: Basomedian tergal light markings present on III-VII.

MALE (allotecnon only). Palpus with basomedian light scales on segments 4 and
a

MALE GENITALIA (allotecnon only). Lobe of tergite IX not prominent, with
3,4 strong bristles. Sidepieces separated at base by a distance less than width of
aedeagus; without any indication of lobes on tergomesal or sternomesal margins.
Claspette represented by a conical tubercle. Aedeagus shorter than proctiger; apex
shallowly emarginate, lateral lobes inconspicuous.

PUPAE. Cephalothorax: Trumpet moderate, index about 6.5, not expanded api-
cally; hairs 10,12-C usually single or double. Abdomen: Hair 5-IV-V long, usually ex-
ceeding apex of second tergite following; hair 9-VII short, slightly larger than 6, usu-
ally not exceeding lateral angle of tergite. Paddle: Moderate to long, usually slightly
produced on apex.

LARVAE. Head: Hair 4-C weak, lightly pigmented, at least 4-branched; 5-C sin-
gle, moderate, strongly pigmented; 6-C heavy, strongly pigmented, double; 7-C mod-
erate, double or triple; 15-C very short, not extending cephalad of base of mental
plate. Thorax: Quadrate in outline; integument glabrous. Abdomen: Integument
covered with small spinelike spicules, restricted to dorsal and ventral surfaces (only
ventral in guerrero). Hair 8-VII caudolaterad of 9-VII. Siphon: Integument imbricate;
acus distinct, separate. Anal Segment: Hair 3-X double; ventral brush with 5 pairs of
hairs.

DISCUSSION. The Allotecnon Group of 3 species is readily recognized by the
unique pleural scale pattern of the adults (particularly the long narrow upper stp
scale patch) and the short, spinelike spicules on the abdominal segments of the lar-
vae.

The group is known only in Central America and is sympatric with the Quadri-
vittatus Group from southern Mexico to central Panama (figs.5,6). A. allotecnon, the
only widespread species, occupies primarily the central and southern parts of the
range of the group but is also known from Chiapas and the southern end of the Sierra
Madre Oriental in Mexico; guerrero is apparently endemic to the Sierra Madre del
Sur in Mexico; and guatemala is known only from 2 collections on the Pacific slope
of central Guatemala. The 3 species appear to be allopatric.

Berlin: Subgenus Howardina 99

24. Aedes (Howardina) allotecnon Kumm, Komp and Ruiz
Figs.6,9 47,48

1940.Aedes (Howardina) allotecnon Kumm, Komp and Ruiz, 1940:417. TYPE: Lectotype d with
associated larval skin, Poas Volcano, Costa Rica, elev. 9000 ft, 14 Apr 1933, H.W. Kumm
[USNM; selection of Stone and Knight 1956:214]. :

Aedes (Howardina) allotecnon of Kumm and Komp (1941:18-22); Kumm and Zuniga (1942:404);
Trapido, Galindo and Carpenter (1955:530,534); Stone and Knight (1956:213); Vargas (1956:
22); Stone, Knight and Starcke (1959:173); Belkin, Schick and Heinemann (1965:12); Forattini
(1965:399); Stone (1967:208).

FEMALE. Wing: 3.5 mm. Proboscis: 2.71 mm. Forefemur: 2.1 mm. Abdomen:
about 3.24 mm. Head: Light decumbent scales on median stripe and short orbital line
creamy; broad light scales on sides and venter whitish. Erect scales on occiput dark.
Palpus about 0.19 of proboscis, with white scales on apex and mesal side of segment
4. Thorax (fig.9): Inner and outer dorsocentral lines creamy to pale golden, com-
plete; lateral marginal line short, extending from scutal angle to wingbase, appearing
broad due to presence of creamy scales in antealar area above paratergite. Median
scutellar lobe with moderately broad dark scales laterally and median creamy to
golden narrow scales from base to apex; lateral lobes with a few narrow creamy
scales. Paratergite with broad white scales. Apn with predominantly broad white
scales, some anterior ones narrow, creamy. Ppn with large upper patch of narrow
creamy scales and small lower patch of broad white scales; psp without scales; other
scale patches silvery white and distributed as in group. Legs: Forefemur predomi-
nantly dark anteriorly except for basal creamy scales, posterior surface mostly cream-
y to middle and with a narrow light line bordered on sides by dark scales extending
distally; midfemur mostly dark anteriorly except for basal creamy scales and a sil-
very spot at distal 0.67, mainly creamy posteriorly, dark scales restricted to distal
half; hindfemur creamy anteriorly to 0.67 and dark distally except for a silvery spot
at 0.75, creamy posteriorly to proximal 0.5, rest dark. Foretarsus with basal white
scales on segment 1; midtarsus with short, incomplete white rings on 1,2; hindtarsus
with moderate white rings on 1-3. Haltere: Stem pale, knob mainly dark, a few
fringe scales whitish. Abdomen: Basomedian tergal light scales white from segment
III distad. Sternites with broad basal V-shaped patches, creamy at base, silvery on
sides; median apical light scales evident to segment IV.

MALE. Wing: 2.62 mm. Proboscis: 2.19 mm. Forefemur: 1 q71 mm. Essentially
similar to female in color. Mesal light scale patch on torus absent.

MALE GENITALIA (fig.47). As figured; diagnostic characters as in the key. Ninth
tergite lobe not prominent, with 3,4 strong setae. Sidepiece length 3.0 of median
width. Clasper about 0.6 of sidepiece, with 2 preapical hairs; spiniform short, about
0.25 of clasper. Aedeagus shorter than proctiger; dorsal sclerotization shallowly emar-
ginate, lateral lobes inconspicuous.

PUPA (fig.47). Abdomen: 3.08 mm. Trumpet: 0.43 mm; index 6.5. Paddle: 0.7
mm. Diagnostic characters as in the key. Chaetotaxy as given below; hairs moderate-
ly to darkly pigmented. Cephalothorax: Mesonotum moderately pigmented between
trumpets, lateral areas and metanotum slightly darker; trumpets contrasting. Hair
1(1), 21, 1-2), 3(1, 1-2), 4(2, 2-4), 5(2), 6(1, 1-2), 7(2), 8C1, 1-3), 91), 10€1, 1-4),
11(1), 12(2, 1-2). Abdomen: Tergite I deeply pigmented, except for usual membra-
nous areas; tergites II and III deeply pigmented, other segments lighter. Segment I:

100 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

hair 1(5-13 primary branches, each with many secondary ones), 2(1, 1-2), 3(1, 1-2),
4(3,.2-5), 5(1; 1-2), 6C1, 1-2), 7C1, 1-2), 901), alveoli of 10,11 present in some. Seg-
ment II: hair 0(1), 1(3, 2-6), 2(1), 31, 1-2), 4(3, 2-5), 5C1), 6C1, 1-2), 7C1, 1-2), 81,
present in a few), 9(1). Segment III: hair 0(1), 1(3, 2-4), 2(1), 3(1), 4(1, 1-2), 5(1),
6(1), 7(2, 1-4), 81, 1-3), 91) 10(1), 11(1), 141). Segment IV: hair 0(1), 1(@, 2-3),
2(1), 3(2, 1-4), 41); SQ, 1-3), 60) 703/14). 80 1-2), 9G), 10), FD, 14C1):
Segment V: hair 0(1), 1(2, 1-3), 2(1), 3(1), 4@, 1-3), 5(2, 1-2), 6(1), 7@, 2-4), 8(1,
1-2), 9(1), 101), 11€1), 14(1). Segment VI: hair 0(1), 1(1, 1-2), 2(1), 31), 4(1),
5(1, 1-2), 6€1), 7C1, 1-2), 8(1,.1-2), 901), 1001), 1101), 14C1). Segment VII: hair 0(1),
ty, 207, 3C1, 1-2), 4Ch 1-4). 51, 1-2), o@2, 1-4), 701), 8, 1-3), 96, 2-7), 10d),
11(1), 14(1). Segment VIII: hair 0(1), 4(1), 9(3, 3-5), 14(1). Paddle: Uniformly light-
ly pigmented, midrib darker; external margin serrated; hair 1-P slender. Male genital
lobe extending to about 0.4 of paddle, female genital lobe to 0.35.

LARVA (fig.48). Head: 0.84 mm. Siphon: 1.00 mm. Anal Saddle: 0.32 mm. Read-
ily separated from guerrero by the presence of abdominal spicules on both dorsal and
ventral surfaces and anal saddle with distinct large marginal spicules ventrad of hair
1-X. Head: Width about 1.06 of length. Pigmentation moderate except around
mouthparts where it is deeper. Mental plate with a strong median tooth and usually
11(11,12) distinct teeth on either side, often a smaller detached tooth at base. Hair
OC1).. 101). 301). 46..4-8), 5C1),6(2),. 7, 2-3), 8C1), 9(3,:3-5), OC, 1-3), t1a12,2-
17), 12(1, 1-2), 131, 1-2), 14(8, 4-9), 15(@2, 2-3). Antenna: Length about.0.37. of
head. Pigmentation moderate; shaft with a few scattered spicules. All hairs except I,
single; hair 1-A (2, 2-4) at about 0.63 from base. Thorax: Apices of long hairs atten-
uate; barbs present on both long and stellate hairs. Prothorax: hair 0(24, 17-29), 1(6,
4-7), 2(1), 3(16, 11-19), 4(2, 1-2), 5(3, 2-3), 6(1), 7(3, 2-4), 8(28, 24-30), 9(28, 22-
28), 10(1), 11(2, 2-4), 12(1), 13(33, 22-35), 14(30, 24-38). Mesothorax: hair 1(11, —
6-14), 2(1), 3(1), 4(1), 51), 6(6, 4-7), 7(1), 8€6, 5-6), 9(7, 5-8), 10(1), 11@, 1-2),
12(1), 13(16, 10-17), 14(8, 8-13). Metathorax: hair 1(16, 12-25), 2(1), 365, 4-9),
4(17, 13-20), 5€15, 13-20), 6(1), 7(6, 3-8), 8(11, 10-14), 9(4, 3-6), 10(1), 111, 1-2),
12(1), 13(27, 20-33). Abdomen: Integument of segment I glabrous; other segments
with spinelike spicules on dorsal and ventral surfaces. Stellate hairs with branches of
unequal length. Segment I: hair 1(19, 12-20), 2(28, 17-28), 3(1), 4(6, 4-8), 5(16, 12-
19), 6(3, 2-3), 71, 1-2), 9(6, 5-9), 10(1),; 11(17, 13-21), 13(10, 8-10). Segment II:
hair 0(1), 1(17, 11-20), 2(26, 17-26), 3(1), 4(6, 4-8), 5(16, 14-20), 6(3, 2-3), 7(9, 6-
11), 8(1, 1-3), 9(6, 4-9), 10(1), 11(1), 12(1), 13(9, 7-16). Segment III: hair 0(1),
1(18, 8-22), 2(24, 16-27), 3(1), 4(1), 5(17, 12-26), 6(2, 2-3), 709, 6-12), 8(1), 9(14,
10-18), 10(1), 111), 12(1), 13(6, 5-12), 14(1). Segment IV: hair 0(1), 1(14, 5-18),
2(23, 15-27), 301,.1-2), 401), 5(15,10-20), 6(@2, 2-3), 7€9, 6-13); 8C1); 9C16, 12-22),
10(1), 11(1), 12(1), 13(7, 4-12), 14(1). Segment V: hair 0(1, 1-2), 1(12, 5-15), 219,
15-25), 3(1), 4C1, 1-3), 5(15, 11-18), 6(2), 7(8, 6-11), 8(1), 9(16, 11-23), 10(1),
11(1), 12(1), 13(8, 4-10), 14(1). Segment VI: hair 0(1), 114, 6-15), 2(17, 13-24),
3(1), 4(1), 512, 10-19), 6(2, 2-3), 7(8, 6-11), 8(1), 9(14, 11-17), 10(1), 11Q1), 121),
13(8, 4-10), 14(1, 1-2). Segment VII: hair 0(1), 1(11, 6-16), 2(16, 11-20), 31), 4(1),
5(1 1 8-17), 6(9, 6-14), 7(1); 86, 3-7), 9013, 10-21), 1001),. 1101), 12), 13(,, 3-
14), 14(1, 1-3). Segment VIII: Comb scales in a single row, usually 14(12-20) in
number; individual scale elongate, pointed apically and evenly fringed nearly to tip.
Hair 0(1), 1(10, 8-20), 2(1), 3(5, 4-6), 4(1), 5(6, 4-8), 14(2, 1-4). Siphon: Index
about 4.0-4.2; integument moderately pigmented, imbricate. Pecten extending to
about 0.55, with 22(17-24) teeth; individual tooth pointed apically, with basal den-
ticles. Hair 1(3, 2-3), 2(1), 3-5 not visible, 6(1), 7(1), 8(1), 9(1). Anal Segment: Sad-

Berlin: Subgenus Howardina : 101

dle imbricate; moderately pigmented, with darker basal ring; caudal margin with
spines of varying sizes dorsad and ventrad of hair 1-X. Gills subequal, about 2.0 of
saddle length. Hair 1(3, 2-4), 2(4, 3-5), 3(2), 4a(4, 2-4), 4b(3, 3-4), 4c(4, 3-4), 4d@3,
2-4), 4e(2, 2-3).

SYSTEMATICS. A. allotecnon is the most widespread and most common of the
3 allopatric species in the Allotecnon Group. Its known range extends from the
southern end of the Sierra Madre Oriental in Mexico to central Panama. The only sig-
nificant variation noted in the large series of specimens was the presence of a few sil-
very scales on upper ppn in the 2 females from Mexico (MEX 14,16). Although this
condition tends toward guerrero the 2 forms appear to be distinct in other adult fea-
tures and the 3 larvae of allotecnon from Mexico are clearly different from those of
guerrero and apparently indistinguishable from those of allotecnon populations in
Central America. Only with additional material of all stages from Mexico will it be
possible to determine the taxonomic status of this population and its relation to
guerrero. The distribution of allotecnon appears to be discontinuous in Central
America probably because of its apparent restriction to areas of cloud forest. Al-
though not reported as yet from Guatemala and Honduras, it may occur there in
suitable environments.

BIONOMICS. The immature stages of allotecnon have been collected in epiphytic
bromeliads at elevations above 5,000 feet, frequently in association with members of
the Quadrivittatus Group. A few females have been taken in daytime biting-landing
collections on human bait. ;

DISTRIBUTION (fig.6). Cordilleras of Middle America in Mexico (Sierra Madre
Oriental and Sierra Madre de Chiapas), possibly eastern Guatemala, El Salvador,
probably Honduras, Nicaragua, Costa Rica and Panama (to Canal Zone) at elevations
above 5000 ft. Material examined: 521 specimens; 43 6, 48 2, 136 pupae, 294 larvae;
114 individual rearings (24 pupal, 46 larval, 44 incomplete).

COSTA RICA. Alajuela: Poasito, 14 Nov 1962, C.L. Hogue and W.A. Powder (CR 23), 5 Ip?
(23-101,102,104,106,110), 1 pd (23-111), 2 p¥ (23- 112,113), 3 IP (23-103,107,108), 6 P, 6 p,
Bone we | [UCLA] . Volcan Poas, 14 Apr 1938, H.W. Kumm, 1 d;1L, 41 [USNM] ; Summit, 14 Nov
1962, C.L. Hogue and W.A. Powder, 1 Ipd (20-110), 1 Ip? (20-109), 6 IP (20-102-106,108), 5 L
(CR 20), 2 Ipd (21-103,106), 2 Ip? (21-109,110), 1 pd (21-112), 1 p? (21-101), 6 IP (21-102,104,
105,107,109,111), 25 L (CR 21); 9 July 1963, C.L. Hogue, 3 6, 2 9,4 P,4p,4L,41(CR 145)
[UCLA]. Cartago: Cervantes, 9 Nov 1962, J.N. Belkin, C.L. Hogue and W.A. Powder, 1 L(CR 9),
3 IP (11-101-103), 2 Ipd (13-104,108), 3 pd (13-102,109,110), 3 p? (13-101-103), 3 IP (13-103,
105,106), 2 L (CR 13), 7 Ipd (14-102-105,107,111,112), 2 lp? (14-109,110), 1 pd (14-101), 1 IP
(14-108), 2 L (CR 14); 9 Nov 1962, C.L. Hogue (CR 12), 1 lpg (12-106), 1 IP (12-105) [UCLA].
La Sierra, elev. 6300-8500 ft, 24 Nov 1962, C.L. Hogue and W.A. Powder, 1 IP (46-201), 1 Ipd
(49-106), 4 Ip? (49-104,108-110), 4 IP (49-101-103,105), 1 p, 1 L, 11(CR 49), 1 pd (51-102),
1 p? (51-101), 2 IP (51-104,105), 10 L (CR 51), 2 Ip (52-101,109), 1 Ip? (52-104), 4 IP (52-102,
103,107,110), 21 L (CR 52), 1 IP (53-104), 1 9, 15 L(CR 54) [UCLA]. Orosi, 7 Dec 1962, CLL.
Hogue and W.A. Powder (CR 83), 1 lp? (83-501), 1 L [UCLA]. Rio Macho, 7 Dec 1962, C.L.
Hogue and W.A. Powder, 1 lp? (79-102), 1 pd (79-101), 2 Ipd (86-103,104), 1 pd (86-101), 1 L
(CR 86) [UCLA]. Volcan Iraju, 15 Nov 1962, C.L. Hogue and W.A. Powder (CR 24), 2 pd (24-
101,108), 1 p? (24-106), 3 IP (24-102-104), 1 P, 1 1 [UCLA]. Heredia: Puerto Viejo, 8 Sept
1965, W.J. Hjort (CR 220), 1 Ipd (220-11) [UCLA]. San Jose: Villa Mills, 21 Nov 1962, C.L.
Hogue and W.A. Powder (CR 39), 1 lp? (39-104), 3 p? (39-101-103), 3 L [UCLA].

EL SALVADOR. Laguna de las Ranas, Jan 1942, H.W. Kumm, 4 6, 1 , 2 L [USNM].

MEXICO. Chiapas: Sumidero Canyon, 17 Aug 1964, E. Fisher and D. Verity, 3 L (MEX 116),
1 ? (MEX 121) [UCLA]. Veracruz: Cordoba, Fortin de las Flores, 11 July 1964, E. Fisher, 1 °
(MEX 14) [UCLA]. Also from states of Hidalgo, Queretaro and San Luis Potosi (Vargas 1956:272).

NICARAGUA. Jinotega, 3 July 1964, J.N. Belkin and A. Quinonez (NI-33C), 1 IP (33C-41)
[UCLA].

102 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

PANAMA. Bocas del Toro: Chiriquisito, Trail Caldera, 5 Nov 1965, P. Orjuela (PA 1004), 6?
[UCLA]. La Zorra, elev. 4400-5400 ft, 6 Apr 1963, A. Quinonez, 1 L (PA 176); 7 Apr 1963, A.
Quinonez, 5 L (PA 179); 8 Apr 1963, A. Quinonez, 4 L, 1 1(PA 180); 9 Apr 1963, A. Quinonez,
7 L (PA 183) [UCLA]. Chiriqui: Palo Santo, elev. 4600 ft, 15 June-1 Dec 1950, 1 2 (Trapido, Ga-
lindo and Carpenter 1955:530). Boquete, 30 Jan 1914, J. Zetek, 1 @ [USNM]. El Hato del Vol-
can, Chiriqui Lagoon, elev. 4500 ft, 12 Mar 1964, A. Quinonez (PA 638), 2 lp? (638-101,102),
1 IP (638-108), 1 L [UCLA]. Cocle: El Valle, elev. 2000 ft, 24 Aug 1963, A. Quinonez (PA 533),
1 p? (533-101), 2 L [UCLA]. Panama: Cerro Campana, 3 Aug 1963, A. Quinonez (PA 462),
1 lp? (462-108), 1 p? (462-101) [UCLA].

25. Aedes (Howardina) guerrero Berlin, n.sp.
Figs.6,9 49

TYPES: Holotype ? (MEX 166), 21 miles south of Chilpancingo, Guerrero, Mexico, 31 Aug
1964, E. Fisher [USNM]. Paratypes: 2 9 (MEX 166), same data as holotype [UCLA].

FEMALE. Wing: 3.06 mm. Proboscis: 2.34 mm. Forefemur: 2.16 mm. Abdomen:
about 3.15 mm. In general similar to allotecnon, differing apparently only in the fol-
lowing. Head: Erect scales on occiput whitish in middle, darker on sides. Palpus about
0.2 of proboscis, with white scales on apex and mesal side of segment 4. Thorax (fig.
9): Mesonotal light lines golden. Creamy scales on median scutellar lobes narrow dis-
tally, broader basally. Paratergite with broad silvery scales. Scales on apn broad, sil-
very; upper ppn with predominantly broad, silvery scales, creamy narrow ones re-
stricted to upper border; lower ppn patch small, with broad silvery scales; psp usual-
ly with a small silvery patch; other scale patches bright silvery. Legs: Foretarsus and
midtarsus with short incomplete white rings on segments 1,2; hindtarsus with broad-
er white rings on 1-3. Haltere: Stem pale, knob mostly creamy, with a few black bas-
al scales. Abdomen: Basomedian tergal light scales white from segment III caudad,
silvery on segments VI,VII. Sternites with broad basal V-shaped white scale patches,
silvery on distal segments; apical white scales inconspicuous.

MALE. Unknown.

PUPA. Imperfectly known. Appears very similar to allotecnon but possibly differ-
entiated by the hair 5-IV,V which is shorter, sorry extending beyond the apex of
the following tergite.

LARVA (fig.49). Head: 0.81 mm. Siphon: 0.7 mm. Anal Saddle: 0.3 mm. Read-
ily separated from allotecnon by the presence of abdominal spicules only on the ven-
tral surface and anal saddle without distinct marginal spicules ventrad of hair 1-X.
Head: Width subequal to length. Pigmentation as in allotecnon. Mental plate with a
strong median tooth and 13(12-14) distinct teeth on either side, often a smaller de-
tached tooth at base. Hair 0(1), 1(1), 3(1), 465, 5-6), 5(1, 1-2), 6(2), 7(2), 8(1), 9Q,
2-4), 10(2, 2-3), 11(7, 5-8), 12(2, 1-2), 13(1), 14(5, 4-8), 15(4, 3-4). Antenna: Length
about 0.33 of head; shaft pigmentation moderate, spicules not apparent. All hairs ex-
cept 1 single; 1-A (2) at about 0.6 from base. Thorax: Integument glabrous as in allo-
tecnon. Apices of long hairs attenuate; barbs present on both long and stellate hairs.
Prothorax: hair 0(13, 11-15), 1(6, 3-7), 2(1), 3(13, 10-14), 4@, 2-3), 5C1, 1-2), 6(2),
~7(2), 8(21, 21-25), 9(17, 13-19), 10(1), 1102, 2-4), 12(1), 1315, 9-17), 1434, 30-
38). Mesothorax: hair 1(7, 4-7), 2(1), 3(1), 41), 5(1), 6(5, 5-6), 7(1), 8(7, 6-7), 9(8,
6-9), 10(1), 11(1), 12(1), 13(14, 10-16), 14(4, 3-7). Metathorax: hair 1(5, 4-6), 2(1),
3(5, 2-6), 4(9, 8-10), 5(9, 8-11), 6(1), 7(7, 6-8), 8, 6-11), 9(4), 10€1), 11€1), 12),

Berlin: Subgenus Howardina 103

13(16, 11-17). Abdomen: Integument of segment I glabrous; other segments with
spinelike spicules restricted to ventral surface; hairs 1,7,13-II-VII apparently not stel-
late. Segment I: hair 1(7, 4-8), 2(11, 10-15), 3(1), 4(4, 4-5), 5(12, 10-13), 6(3, 3-4),
7(1), 9(6, 5-9), 10(1), 11€11, 10-15), 13(5, 4-7). Segment II: hair 0(1), 1(4, 4-7),
2(9, 8-12), 3(1), 4(4, 4-6), 5(11, 9-12), 6(3, 3-4), 7(5, 4-7), 8(1), 9(5, 5-7), 10(1),
11(1), 12(1), 13(4, 4-5). Segment III: hair 0(1), 1(4, 3-5), 2(11, 8-11), 3(1), 4(1),
5(11, 9-12), 6(2), 7(5, 4-6), 8(1), 9(7, 5-9), 10(1), 111), 121), 13(4, 3-5), 14(1).
Segment IV: hair 0(1), 1(4, 3-5), 2(10, 7-11), 3(1), 41), 510, 10-13), 6(2), 75, 5-
7), 8(1), 9(8, 6-8), 10(1), 11(€1), 12(1), 13(4, 4-5), 14(1). Segment V: hair 0(1), 1(4,
3-4), 2(8, 8-10), 3(1), 4(1, 1-2), 5(10, 8-11), 6(2), 7(5, 5-8), 8(1), 9(7, 5-9), 10(1),
11(1), 121), 1305, 3-6), 14(1). Segment VI: hair 0(1), 1(3, 3-5), 2(8, 5-9), 31),
4(1), 5(9, 6-11), 6(2), 7(4, 3-5), 8(3, 3-4), 9(7, 6-9), 10(1), 11€1), 121), 13(5, 4-5),
14(1). Segment VII: hair 0(1), 1(3, 3-4), 2(8, 7-10), 3(1), 41), 5(7, 5-9), 6(3, 2-4),
7(1, 1-2), 8(5, 5-7), 9(5, 3-9), 10(1), 111), 12(1), 13, 3-5), 14(1). Segment VII:
Comb scales in a single row, usually 14(12-16) in number; individual scale elongate,
pointed apically, evenly fringed almost to tip. Hair 0(1), 1(9, 4-11), 2(1), 3(7, 6-9),
4(1), 5(6, 5-7), 14(1). Siphon: Index about 3.3-3.6. Integument imbricate, moder-
ately to strongly pigmented, with darker basal ring. Pecten extending to 0.42, usual-
ly with 18(17-22) teeth; individual tooth pointed apically, with basal denticles. Hair
1(3, 2-4), 2(1), 3-5 not visible, 6(1), 7(1), 8(2, 1-2), 9(1). Anal Segment: Saddle im-
bricate, moderately to strongly pigmented, with darker basal ring; caudal margin
with spicules present only dorsad of hair 1-X. Gills subequal, long, more than 2.0 of
saddle. Hair 1(4, 3-5), 2(5, 5-7), 3(2), 4a(4, 3-5), 4b(3, 3-4), 4c(3, 3-4), 4d(3, 2-3),
4e(2, 1-2).

SYSTEMATICS. Although closely resembling allotecnon, this form appears to be
a distinct species restricted to the Sierra Madre del Sur in Mexico where no other
member of the group has been collected. The association of the immature stages with
the females is presumptive only as no adults were obtained from the rearings and
were found in a different locality from the adults but is probably correct since the
larva is obviously of the allotecnon type but with distinctive features as indicated in
the diagnosis. The pupal stage is not described at this time as it is known only from
2 whole imperfect specimens from incomplete rearings. It also appears to be general-
ly similar to allotecnon but possibly differentiated by the shorter hair 5-IV-V.

BIONOMICS. The presumptive immature stages of guerrero have been collected
in epiphytic bromeliads at elevations of about 7,200-7,800 feet (MEX 429,430,431,
434). The 3 females were taken in a biting-landing collection (MEX 166) at 0800 in
a pine-oak woodland.

DISTRIBUTION (fig.6). Sierra Madre del Sur in Mexico at elevations above 7000
ft. Material examined: 20 specimens; 3 9, 2 pupae, 15 larvae; 2 incomplete individual
rearings.

MEXICO. Guerrero: Chilpancingo, 21.0 mi S, 21 Aug 1964, E. Fisher, 3 ? (MEX 166, type se-
ties) [USNM, UCLA]. Oaxaca: Ixtlan de Juarez, El Punto, elev. 7700 ft, 13 Aug 1966, D.A. Schroe-
der (MEX 434), 1 IP (434-10); Vivero Rancho Teja, elev. 7200 ft, 12 Aug 1966, D.A. Schroeder,
1 IP (429-10), 7 L (MEX 429), 1 L (MEX 430), 5 L (MEX 431) [UCLA].

26. Aedes (Howardina) guatemala Berlin, n.sp.
Figs.6,9

TYPES: Holotype ? (GUA 138-13), Yepocapa, Chimaltenango, Guatemala, 20 July 1950, Mi-

104 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

guel Xinic and Pedro Garcia [USNM]. Paratypes: 4 9? (GUA 138-10,12,14,15), same data as holo-
type [UCLA].

FEMALE. Wing: 3.2 mm. Proboscis: 2.37 mm. Forefemur: 2.01 mm. Abdomen:
about 2.6 mm. Very similar to allotecnon, differing apparently only in the following.
Head: Erect scales on occiput predominantly dark, median ones lighter. Thorax (fig.
9): Mesonotal light lines pale golden. Lateral scutellar lobe with a few narrow dark
and pale golden scales. Upper ppn scales predominantly creamy, narrow, a few scales
slightly broader at lower border. Legs: Foretarsus and midtarsus with a few white
scales on segment 1; hindtarsus with short white rings on segments | and 2 only.
Haltere: Stem pale, knob predominantly white, dark scales restricted to base.

MALE, PUPA, and LARVA unknown.

SYSTEMATICS. The 6 known females of guatemala are so different in the de-
tails of ornamentation from other members of the Allotecnon Group that I do not
hesitate in recognizing them as members of a distinct species even in the absence of
immature stages. This species appears to be restricted to the Pacific slope of central
Guatemala where no other species of Howardina has been recorded. _

BIONOMICS. The immature stages will probably be found in epiphytic bromeli-
ads. The females were probably taken in a biting-landing collection on human bait.

DISTRIBUTION (fig.6). Pacific slope of Guatemala. Material examined: 6 speci-
mens; 6 2; no individual rearings or immature stages.

GUATEMALA. Chimaltenango: Yepocapa, 20 July 1950, Miguel Xinic and Pedro Garcia (GUA
138, type series), 5 2 (138-10,12,13-15); 24 July 1950, V.C. and R. Feoli, 1 @ (GUA 139) [USNM,
UCLA].

Eleanorae Group

FEMALES. Head: Median light line on vertex usually with narrow scales (moder-
ately broad in eleanorae). Proboscis distinctly longer than forefemur. Torus with a
restricted mesal scale patch. Thorax: Mesonotal light lines usually creamy to golden
(except in ecuadoriensis in which inner dorsocentral and lateral marginal lines are
white); inner dorsocentral line complete; outer dorsocentral line usually shorter, not
reaching anterior margin, subparallel to inner; lateral marginal line continuous; scales
on anterior promontory creamy, golden or white. Antealar area above paratergite
with golden scales. Paratergite with a moderate scale patch extending entire sclerite
length. Upper ppn entirely covered with light scales; upper stp patch quadrate. Legs:
Distal white spot on anterior surface of midfemur distinct; present or absent on hind-
femur. Tibiae completely dark. Wing: Veins entirely dark scaled. Abdomen: Basome-
dian tergal light markings absent or present on III-VII.

MALES. Palpus with or without basomedian light scales on segments 4 and 5.

MALE GENITALIA. Lobe of tergite IX prominent or indistinct, with 4-7 strong
bristles. Sidepieces separated at base by a distance less than width of aedeagus, with-
out any indication of lobes on tergomesal or sternomesal margins. Claspette repre-
sented by a conical tubercle. Aedeagus shorter than or subequal to proctiger; apical
emargination distinct or shallow, lateral lobes conspicuous or inconspicuous.

PUPAE. Cephalothorax: Trumpet moderate, index 4.5-6.4, not expanded apical-
ly; hairs 10,12-C varied in branching. Abdomen: Hair 5-IV-V moderate to long, some-
times exceeding apex of second tergite following; hair 9-VII moderate, reaching or
exceeding lateral angle of tergite. Paddle: Moderate; apex usually rounded.

Berlin: Subgenus Howardina | 105

LARVAE. Head: Hairs 4-7-C varied in branching; 15-C moderate to long, always
extending beyond base of mental plate. Thorax: Roundish in outline; spiculation
dense, weak or absent, spicules hairlike, simple. Abdomen: Integument strongly or
weakly uniformly spiculate. Hair 8-VII distinctly cephalodorsad of 9-VII. Siphon: In-
tegument smooth or imbricate; acus distinct, separate. Anal Segment: Hair 3-X single
or double; ventral brush with 5 pairs of hairs.

DISCUSSION. The Eleanorae Group of 6 species is probably derived from the
Quadrivittatus Group and seems to replace the Allotecnon Group south of Panama.
The ornamentation of the midfemur and hindfemur is quite varied in this group,
more or less bridging the characteristic development in the other 2 groups. The lar-
vae are also quite varied and somewhat intermediate between the Quadrivittatus and
Allotecnon Groups, although the ventral brush always has only 5 pairs of hairs as in
the latter group. Members of this group have been misidentified in the past as quad-
rivittatus.

The Eleanorae Group appears to be entirely confined to the Andes where it has
been collected from moderate elevations up to nearly 11,000 feet. The group is now
known from western and possibly northcentral Venezuela, Colombia, Ecuador and
Bolivia (fig.6). Although seemingly some of the species are found sympatrically,
there are indications of strong ecological divergence in other instances (ecuadoriensis
and brevivittatus for example). This group is apparently sympatric to a limited ex-
tent with the Whitmorei Group in the north and with the Whitmorei and Aurivitta-
tus Groups in the south.

27. Aedes (Howardina) eleanorae Berlin, n.sp.
Figs.6,9,50,5 1

TYPES: Holotype ¢ (COB 10-21) with associated pupal and larval skins and genitalia slide,
Buritica, Antioquia, Colombia, 5000 ft, 8 Jan 1965, E. Osorno-Mesa [USNM]. Allotype ° (COB 10-
20) with associated pupal and larval skins, same data as holotype [USNM]. Paratype ? (COB 10-
22) with associated pupal and larval skins, same data as holotype [UCLA]. This species is dedicated
to my wife, Eleanor, for her interests in and contribution to the study of this group.

Aedes (Howardina) quadrivittatus in part of Dyar (1928:229); Edwards (1932:155); Lane (1939:
100; 1953:713); Reyes (1955:74,89); Stone, Knight and Starcke (1959:174).

FEMALE. Wing: 3.06 mm. Proboscis: 2.34 mm. Forefemur: 2.07 mm. Abdomen:
2.88 mm. Head: Scales on median stripe predominantly broad and white, a few ante-
rior ones narrow; light scales on sides and venter white. Erect scales on occiput and
sides creamy. Palpus about 0.16 of proboscis. Mesal light scale patch on torus creamy.
Thorax (fig.9): All mesonotal light lines creamy, of uniform width throughout; outer
dorsocentral line short, arising at anterior 0.25 and reaching scutellum; lateral mar-
ginal line long, from side of anterior promontory to wingbase, appearing broad from
scutal angle to wingbase owing to presence of creamy scales in antealar area above
paratergite; scales on anterior promontory creamy. Median scutellar lobe with broad
dark scales laterally and broad creamy scales in the middle from base to apex; lateral
lobes with a few narrow creamy scales. Scales on paratergite moderately broad, white.
Apn scales broad, white, a few anterior ones narrower, creamy; upper ppn predomi-
nantly with narrow, curved scales, a few broader scales below; lower ppn patch dis-
tinct and merging with ssp patch; other scale patches pure white, distributed as in the

106 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

group. Legs: Forefemur dark anteriorly, predominantly dark posteriorly except for a
creamy longitudinal line; midfemur with narrow anteromesal creamy line to 0.67
terminating in a white spot, posteriorly proximal 0.5 creamy, rest dark; hindfemur
creamy to 0.67 on both sides, rest dark scaled. Foretarsus with a few white scales on
segment 1; midtarsus with short incomplete white ring on 1; hindtarsus with broader
white rings on 1 and 3, second segment dark. Haltere: Stem pale, knob predominant-
ly white, a few dark scales dorsally. Abdomen: Basomedian tergal light scales absent.
Sternites with broad V-shaped white patches; apical light scales inconspicuous.

MALE. Wing: 2.88 mm. Proboscis: 2.43 mm. Forefemur: 1.89 mm. Essentially
similar to female in color. Palpus entirely dark.

MALE GENITALIA (fig.50). As figured; diagnostic characters as in the key. Ninth
tergite lobe not prominent, with 4,5 strong bristles. Sidepiece length about 3.0 of
median width. Clasper about 0.5 of sidepiece, with 1,2 subapical hairs; basal 0.3 with
minute spicules; spiniform long, about 0.5 of clasper. Aedeagus shorter than proc-
tiger; dorsal sclerotization emarginate apically, producing lateral sclerotized lobes.

PUPA (fig.50). Abdomen: 3.3 mm. Trumpet: 0.39 mm; index 4.4-4.5. Paddle:
0.67 mm. Readily separated from other members of the group (except Jeei) by the
strongly developed hair 8-VII on dorsal surface; from Jeei by the branched hair 10-C.
Chaetotaxy as given below, hairs moderately to darkly pigmented. Cephalothorax:
Mesonotum lightly pigmented between trumpets, lateral areas and metanotum dark-
er; trumpet moderately pigmented and sharply contrasting. Hair 1(1), 2(1), 3(1, 1-2),
4(3, 2-5), 5(2), 6(1, 1-2), 7(2), 81, 1-3), 9(1, 1-2), 10(6, 5-6), 11(1, 1-2), 12(4, 2-5).
Abdomen: Tergite I moderately pigmented except for usual membranous areas; II-
VII moderately pigmented, progressively lighter caudad. Segment I: hair 1(13-16 pri-
mary branches), 2(1, 1-3), 3(1), 4(2, 2-4), 5(1, 1-2), 6(1), 71, 1-2), 9(1, 1-3), 11Q,
2-3, present in some). Segment II: hair 0(1), 1(6, 4-7), 2(¢1), 3(1), 4(2, 2-3), 5(2, 1-
2), 6(1), 7(2, 1-3), 9(1), 10 (present in a few). Segment III: hair 0(1), 1(2, 2-4), 2(2,
1-3), 31, 1-2), 4C1, 1-2), 5(1, 1-2), 6(1), 71, 1-2), 8(1, 1-3), 9(1), 10€1), 111),
14(1). Segment IV: hair O(1), 1(2, 2-3), 2(1, 1-2), 3(3, 2-3), 4(1, 1-2), 5(2), 6(1),
71, 1-2), 8(1, 1-2), 9C1), 101), 11€1), 14(1). Segment V: hair 0(1), 1(2), 201, 1-2),
3(2, 1-2), 4(2, 2-3), 5(2), 6(1), 7(2, 1-3), 8(@2, 1-3), 9(1), 10(1), 111), 141). Seg-
ment VI: hair O(1), 1@, 1-2), 2(@2, 1-2), 3(1, 1-2), 4(2, 1-2), 5(1), 6(1), 71, 1-2), 8,
F279. C1), FOLT), (1), P4C1) Segment Vi: hair OC), 1(1, 1-2), 201,142). 3,12),
4(1), 51, 1-3), 6(3, 1-3), 7(1), 8(4, 3-5), 9(4, 4-8), 10(1), 11(1), 14(1). Segment
VIII: hair 0(1), 4(1), 9(11, 11-14), 14(1). Paddle: Lightly pigmented, midrib slightly
darker; hair 1-P single. Male genital lobe extending to 0.5 of paddle, female genital
lobe to 0.34.

LARVA (fig.51). Head: 0.63 mm. Siphon: 0.28 mm. Anal Saddle: 0.27 mm. Diag-
nostic characters as in the key. Readily separated from other members of the group
by pecten teeth extending beyond hair 1-S. Head: Width subequal to length; pigmen-
tation moderate except around mouthparts where it is deeper. Mental plate with a
strong median tooth and usually 12 distinct teeth (11-12) on each side, often a
smaller detached tooth at base. Hair 0(1), 1(1), 3(1), 4(12, 9-15), 5(8, 6-9), 6(5, 4-6),
7(17, 14-18), 8(12, 10-14), 9(14, 13-16), 10(2, 2-3), 11(34, 20-42), 12(10, 7-10),
13(1), 14€18, 18-22), 15(2, 2-3). Antenna: Length about 0.5 of head; shaft uniform-
ly lightly pigmented and with few scattered spicules; hair 1(3, 2-3), others single.
Thorax: Integument densely covered with hairlike simple or branched spicules except
on intersegmental areas; long hairs barbed and with attenuate apices; stellate hairs
with many weakly barbed branches. Prothorax: hair 0(30, 27-34), 1(5, 3-5), 2(1),
3(42, 35-42), 4(2, 1-2), 5(1), 6(1), 7(2), 8(30, 28-34), 9(22, 18-24), 10(1), 112, 2-

Berlin: Subgenus Howardina 107

3), 12(1), 13(36, 31-36), 14(52, 48-62). Mesothorax: hair 1(27, 23-32), 2(1), 3(1),
4(1), 5(1), 6(4, 3-5), 7(1), 8(4, 4-6), 9(5, 5-7), 10(1), 1101, 1-2), 12(1), 13(22, 18-
23), 14(22, 18-26). Metathorax: hair 1(37, 34-40), 2(1), 3(23, 20-26), 4(22, 15-25),
5(21, 17-23), 6(1), 7(4, 4-5), 8(23, 23-35), 9(3, 3-4), 10(1), 111, 1-3), 12(1), 13(26,
24-30). Abdomen: Integument and stellate hairs as on thorax. Segment I: hair 1(28,
25-28), 2(28, 26-30), 3(1), 4(1), 5(24, 22-25), 6(3, 2-3), 7(1), 9(12, 10-15), 10(1),
11(24, 20-24), 13(30, 29-35). Segment II: hair 0(1), 1(30, 29-31), 2(30, 24-32), 3(1),
4(1), 5(29, 25-30), 6(3), 7(20, 19-29), 8(1), 9(17, 16-17), 10(1), 11(1), 12(1), 13(36,
30-40). Segment III: hair 0(1), 1(30, 28-32), 2(31, 26-34), 3(1), 4(1), 5(28, 25-30),
6(2), 7(22, 17-23), 8(1), 9(22, 21-25), 10(1), 11), 12(1), 13(30, 30-33), 14(1). Seg-
ment IV: hair 0(1), 1(32, 27-32), 2(29, 26-35), 3(1), 4(1), 5(30, 24-32), 6(2), 7(22,
18-22), 8(1), 9(26, 22-27), 10(1), 11¢€1), 12(1), 13(28, 26-29), 14(1). Segment V:
hair O(1), 1(28,28-31),:2(30, 28-33), 3(1), 4€1), 5(30, 23-31); 6(2), 7022, 16-25),
8(1), 9(26, 25-26), 10(1), 11(1), 12(1), 13(30, 29-33), 14(1). Segment VI: hair 0(1),
1(30, 30-32), 2(28, 28-33), 3(1), 41), 5(30, 27-30), 6(1), 7(20, 12-21), 8(1), 9(26,
22-26), 10(1), 11(1), 12(1), 13(26, 23-38), 14(1). Segment VII: hair 0(1), 1(30, 28-
33), 2(30, 24-30), 3(1), 4(1), 5(29, 24-29), 6(14), 7(1), 8(24, 20-25), 9(24, 18-24),
101, 1-2), 11€1), 12(1), 13(28, 24-28), 14(1). Segment VIII: Comb scales in a single
row, usually 12(12-14) in number; individual scale long, pointed apically and fringed
almost to tip. Hair 0(1), 1(33, 28-34), 2(1), 309, 8-9), 4(1), 5(32, 27-35), 14(1). Si
phon: Index about 3.3; integument imbricate, moderately pigmented except for
darker basal ring. Pecten teeth usually 21(19-23), extending beyond hair 1-S to
about 0.7 of siphon; individual tooth with basal denticles. Hair 1(4, 3-5), 2(1), 3-5
not visible, 6(1), 7(1), 8(1, 1-2), 9(1). Anal Segment: Saddle moderately uniformly
pigmented except for darker basal ring; integument with minute spicules; caudal mar-
gin with spines of varying length. Gills subequal, moderately long, about 2.0 of sad-
dle length, narrow. Hair 1(2, 2-3), 2(4, 3-4), 3(1), 4a(2), 4b(2), 4c(2), 4d(2), 4e(1).

SYSTEMATICS. A. eleanorae is the most strongly differentiated member of the
group as indicated in the diagnosis above and in the keys. At present it is known
from only 2 localities in Colombia, Buritica in the northern part of the Cordillera
Occidental, and Buena Vista in the upper part of the Orinoco basin on the eastern
slope of the Cordillera Oriental. Although eleanorae may be broadly sympatric with
osornoi and marinkellei, it has not been found associated with these species and ap-
parently occurs at lower elevations.

BIONOMICS. The breeding sites of eleanorae are in epiphytic bromeliads at ele-
vations of about 5,000 feet (COB 10). The female from Buena Vista was probably
taken in a biting-landing daytime collection.

DISTRIBUTION (fig.6). Upper Magdalena and Orinoco drainages in Colombia.
Material examined: 15 specimens; | 6, 3 9, 4 pupae, 7 larvae; 3 individual larval rear-
ings).

COLOMBIA. Antioquia: Buritica, elev. 5000 ft, 8 Jan 1965, E. Osorno-Mesa (COB 10, type se-
ries), 1 Ipd (10-21), 2 Ip? (10-20,22); 15 Jan 1965, E. Osorno-Mesa, 1 p, 1 L, 31 [USNM, UCLA].
Meta: Villavicencio, Buena Vista, 4 June 1942, W.H.W. Komp, 1 9? [UCLA].

28. Aedes (Howardina) marinkellei Berlin, n.sp.
Figs.6,9,52,53

TYPES: Holotype ? with associated pupal and larval skins (COM 63-19), Suba, Bogota, Cun-
dinamarca, Colombia, 8300 ft, 16 Apr 1966, C.J. Marinkelle [USNM]. Paratypes: 1 9 (63-15) with

108 - Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

associated pupal and larval skins, 7 larvae, same data as holotype [UCLA]. This species is dedi-
cated to Dr. C.J. Marinkelle for his contributions to the study of neotropical mosquitoes.

Aedes (Howardina) quadrivittatus in part of Dyar (1928:229); Edwards (1932:155); Lane (1939:
100; 1953:713); Reyes (1955:74,89); Stone, Knight and Starcke (1959:174).

FEMALE. Wing: 3.85 mm. Proboscis: 3.06 mm. Forefemur: 2.19 mm. Abdomen:
about 3.41 mm. Similar to eleanorae, differing chiefly in the following. Head: Light
scales on median stripe and short orbital line narrow, creamy; light scales on sides
and venter broad, creamy. Erect scales predominantly creamy, darker on sides. Pal-
pus about 0.16 of proboscis; predominantly dark, apex and mesal surface of segment
4 white scaled. Mesal patch on torus creamy. Thorax (fig.9): All mesonotal light lines
creamy; outer dorsocentral line complete, broader on anterior 0.5, merging with lat-
eral marginal line at level of lateral prescutal area; lateral marginal line long, moder-
ately broad, extending from area of humeral bristle to wingbase; scales on anterior
promontory creamy. Light scales on median scutellar lobe broad, white from base to
apex; lateral lobes with a few narrow dark scales. Scales on paratergite moderately
broad, creamy. Apn scales predominantly broad, dingy white, a few anterior ones nar-
rower; ppn with only an upper patch of predominantly narrow creamy scales, a few
lower ones whitish, broader; upper stp patch larger than in eleanorae; other scale
patches white, distributed as in eleanorae. Legs: Forefemur predominantly dark an-
teriorly except for a few scattered whitish scales along whole length, posterior sur-
face mostly creamy except for a narrow dark line from base to apex; midfemur pre-
dominantly dark anteriorly except for a few basal creamy scales and a white spot at
distal 0.67, posteriorly mostly creamy with scattered dark scales on distal 0.5; hind-
femur creamy to 0.67 on both sides, rest dark scaled. Foretarsus and midtarsus with
basal white scales on segments 1,2; hindtarsus with broader white rings on segments
1-3. Haltere: Stem whitish, knob predominantly white, a few basal scales dark. Ab-
domen: Basomedian tergal light scales dingy white. Sternites with large basal patches
of dingy white scales; median apical white scales distinct to VI, on segment VII few
and indistinct. Cercus with a few apical white scales.

MALE unknown.

PUPA (fig.52). Abdomen: 4.6 mm. Trumpet: 0.51 mm; index about 4.5. Paddle:
0.86 mm. Chaetotaxy as given below, hairs moderately to darkly pigmented. Distin-
guished from osornoi by metanotal hairs all single, from /eei and eleanorae by ven-
trally placed hair 8-VII. Cephalothorax: Mesonotum moderately pigmented between
trumpets, lateral areas and metanotum darker; trumpets strongly contrasting. Hair
11), 21), 3(€1), 4(3), 5(2), 6(1), 7(2, 2-3), 8(1), 9(1), 10(1), 111), 121). Abdo-
men: Tergite I moderately pigmented except for usual membranous areas; rest of seg-
ments moderate, pigmentation progressively lighter caudad. Segment I: hair 1(6-7
primary branches, each with many secondary ones), 2(1), 3(1), 4(@, 1-2), 5(2), 6(1),
71), 91), 111). Segment II: hair 0(1), 12, 1-2), 2€1), 3C1), 4C1), 51), 61), 701),
9(1).. Segment LI: hair 0(1), 1(1,.4-2), 261), 3C1), 4€1,: 1-2); 5G), 6€1); 71, 1-2),
8(1), 9(1), 10(1), 11(1), 14(1). Segment IV: hair 0(1), 1(1, 1-2), 2(1), 3(2), 4Q, 1-2),
5(2), 6(1), 71, 1-2), 8(1), 91), 10€1), 11€1), 141). Segment V: hair 0(1), 1(1, 1-2),
2(1), SU), 41), S€1); 601): FO), SUT), OCT), TOG). 1101), T4C1).. Segment VT: hair
O(1), 1€1), 2(1), 31), 4€1), 51), 6(1), 701), 8C1), 91), 10€1), 11¢1), 14€1). Segment
VII: hair 0(1), 1(1), 21), 3(1), 41), 5(1), 6(1), 7(1), 81), 9(@2), 10(1), 111), 14(1).
Segment VIII: hair 0(1), 4(1), 9(5, 5-7), 14(1). Paddle: Uniformly lightly pigmented,
midrib darker; external margin serrated; apex rounded; hair 1-P slender. Female geni-
tal lobe extending to 0.31 of paddle.

Berlin: Subgenus Howardina 109

LARVA (fig.53). Head: 0.93 mm. Siphon: 0.81 mm. Anal Saddle: 0.32 mm. Chae-
totaxy as given below, hairs moderately to darkly pigmented. Readily separated from
all other members of Eleanorae Group by hair 5-C always double. Head: Width sub-
equal to length; pigmentation moderate except around mouthparts where it is deeper.
Mental plate with a strong median tooth and 10,11 distinct teeth on either side, of-
ten a smaller detached tooth at base. Hair 0(1), 1(1), 3(1), 4(6, 5-6), 5(2), 6(3, 3-4),
7(5, 4-6), 8(1), 9(2, 2-4), 10(2), 11(12, 8-12; stellate, branches of varied lengths),
12(1), 13(1), 14(6, 4-8), 15(2). Antenna: Length about 0.33 of head; pigmentation
moderate; shaft with scattered spicules. All hairs except 1 single; hair 1-A (2, 2-3) at
about 0.58 from base. Thorax: Integument sparsely covered with simple, hairlike
spicules. Apices of long hairs attenuate; long and stellate hairs weakly barbed. Pro-
thorax: hair 0(17, 13-22), 1(3), 21), 3(20, 19-22), 4(2, 1-2), 5(1), 6(1), 7(2), 8(15,
10-16), 9(11, 9-13), 10(1), 111, 1-3), 121), 13(17, 16-20), 14(34, 34-38). Mesotho-
rax: hair 1(7, 6-8), 2(1), 3(1), 4(1), 5(1), 6(4, 3-5), 7(1), 8(6, 5-6), 9(6, 4-8), 10(1),
11(1), 12(1), 13(10, 8-11), 14(2, 1-3). Metathorax: hair 1(9, 6-9), 2(1), 3(3, 2-4),
4(9, 8-10), 5(10, 7-10), 6(1), 7(6, 5-7), 809, 8-10), 9(3), 10(1), 11€1),.12(1), 1312,
10-18). Abdomen: Integument and stellate hairs as in thorax. Segment I: hair 1(13,
£1517), 2(6, 8-11), 31), 401), 310, 9-16), 663, 2-3), 71), 9: 2-3), 1061); 1133.
11-14), 13(6, 4-8). Segment II: hair 0(1), 1(16, 12-16), 2(9, 8-11), 3(1), 4(1), 5(10,
8-12), 6(3), 7(6, 5-6), 8(1), 9(6, 5-7), 10(1), 11(1), 12(1), 13(9, 8-10). Segment III:
hair O(1), 1(15, 11-16), 210, 10-12), 3(1), 4(1), 5(13, 11-14), 6(2), 7(8, 7-10), 8(1),
9(11, 6-11), 101), 11€1), 12(1), 1311, 8-12), 14(1). Segment IV: hair 0(1), 1(15,
F3-17); 2(12; 9-13), Stl), 401), 3012, 10213); 6); 7, 70), B01). Stl 1-9-1 9).
10(1), 11€1), 121), 13€11, 8-12), 14(1). Segment V: hair 0(1), 1(14, 12-16), 2(10,
£0-13),73(1);-4¢1), 52, 9-13), 00), 70769); Bly; OCT 1-15), 1001), Litt
12(1), 13(10, 9-11), 14(1). Segment VI: hair 0(1), 1(14, 11-16), 2(10, 9-12), 3(1),
4(1), 5(11, 10-13), 6(2), 765, 5-8), 8(1), 9(12, 10-14), 10(1), 11(1), 121), 1309, 7-
9), 14(1). Segment VII: hair 0(1), 1(10, 9-15), 2(10, 10-13), 3(1), 4(1), 5(10, 8-11),
6(5, 4-6), 7(1), 8(5, 5-6), 9(8, 5-10), 10(1), 11(1), 12(1), 13(5, 4-6), 14(1). Segment
VII. Comb scales in a single row, usually 14(14-18) in number; individual scale long,
pointed apically, fringed nearly to tip. Hair 0(1), 1(7, 7-11), 2(1), 3(5, 3-6), 4(1), 56,
6-9), 14(1). Siphon: Index about 3.0-3.2; moderately pigmented, imbricate. Pecten
extending to about 0.5, usually with 16(16-19) teeth; individual tooth with basal
denticles. Hair 1(1, 1-2), 2(1), 3-5 not visible, 6(1), 7(1), 8(1), 9(1). Anal Segment:
Saddle moderately pigmented, with darker basal ring; caudal margin with spines of
varying sizes. Gills subequal, at least 3.0 of saddle length, fingerlike. Hair 1(1), 2(4,
3-4), 3(2, 2-3), 4a(2, 2-3), 4b(2), 4c(2, 1-2), 4d(2, 1-2), 4e(2).

SYSTEMATICS. A. marinkellei is most similar to osornoi and leei but is readily
separated from both by constant diagnostic features in all known stages. It is known
at present only from Cundinamarca Province in Colombia at elevations of over 8,000
feet where it is sympatric with osornoi. As indicated under the latter, a single reared
female from this area combines features of osornoi in adult ornamentation and of
marinkellei in the immature stages, suggesting hybridization between the 2 forms.
Although found in the same general locality, osornoi and marinkellei have not been
collected together in the same bromeliad axils. It appears, therefore, that some eco-
logical barrier usually separates the 2 species.

BIONOMICS. Immature stages of marinkellei have been found in the axils of epi-
phytic bromeliads at elevations above 8,000 feet. Judging by the adult collection, fe-
males of this species are apparently at least occasionally attracted to humans.

DISTRIBUTION (fig.6). Cordillera Oriental in Colombia. Material examined: 22

110 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

specimens; 11 9, 2 pupae, 9 larvae; 2 larval individual rearings. |

COLOMBIA. Cundinamarca: Bogota, 1918, Fr. Apollinaire-Marie, 1 2; B.G. Amortegui, 1
[USNM]. Rio San Francisco, 21 May 1942, W.H.W. Komp, 7 ° [USNM, UCLA]. Suba, elev. 8300
ft, 16 Apr 1966, C.J. Marinkelle (COM 63, type series), 2 lp? (63-15,19), 7 L [USNM, UCLA].

29. Aedes (Howardina) leei Berlin, n.sp.
Figs.6,9,54,55

TYPES: Holotype 2 (COL 17), 17.5 miles S.E. of Purace, Huila, Colombia, 10,000 ft, 28 Oct
1964, V.H. Lee [USNM] . Paratypes: 32 ? (COL 17), same data as holotype [USNM, BM, UCLA].
This species is dedicated to Dr. V.H. Lee, in recognition of his contributions to the study of neo-
tropical mosquitoes.

Aedes (Howardina) quadrivittatus of Martinez and Prosen €1953:28,30; 1955:33-33; 1959:603,
605); Martinez, Carcavallo and Prosen (1960:25); Reyes (1955:74,89); Prosen, Carcavallo and
Martinez (1964:105). :

FEMALE. Wing: 5.04 mm. Proboscis: 3.15 mm. Forefemur: 2.88 mm. Abdomen:
about 4.32 mm. Similar to eleanorae but larger and differing primarily in the follow-
ing. Head: Light scales on median stripe and short orbital line narrow, pale golden;
light scales on sides and venter creamy. Erect scales dull creamy. Palpus about 0.17
of proboscis; predominantly dark, apex and mesal surface of segment 4 creamy.
Mesal patch on torus creamy. Thorax (fig.9): All mesonotal light lines golden; outer
dorsocentral line incomplete, extending from anterior 0.25 to scutellum; lateral mar-
ginal line long, extending from side of anterior promontory to wingbase; scales on
anterior promontory golden. Light scales on median scutellar lobe narrow, golden;
lateral lobes with a few narrow dark scales. Scales on paratergite broad, creamy. Apn
scales predominantly broad, creamy, a few anterior ones narrower, pale golden; ppn
with only an upper patch of predominantly narrow golden scales; a few lower ones
broader; other patches dingy white, distributed as in eleanorae. Legs: Forefemur dark
anteriorly except for basal creamy scales, posterior surface with a creamy line broad
to 0.5, narrowed distally, rest dark; midfemur mostly dark anteriorly except for basal
creamy scales and a dingy white spot at distal 0.67, posteriorly proximal 0.6 creamy,
rest dark; hindfemur creamy to 0.67 on both sides, rest dark scaled. Foretarsus with
a few white scales on segment 1; midtarsus with short incomplete white rings on 1,2;
hindtarsus with broader white rings on segments 1-3. Haltere: Stem pale, knob most-
ly dark, fringe scales whitish. Abdomen: Basomedian tergal light scales distinct; din-
gy white on segments II-VI, white on VII. Sternites with basal creamy patch; VI,VII
with subapical band of white scales; apical light scales inconspicuous.

MALE. Wing: 5.04 mm. Proboscis: 3.69 mm. Forefemur: 2.7 mm. Essentially sim-
ilar to female in color. Palpus predominantly dark, with basoventral white scales on
segments 4 and 5. :

MALE GENITALIA (fig.54). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with 5,6 strong setae. Sidepiece length about 3.0 of median
width. Clasper about 0.62 of sidepiece, with 2,3 subapical hairs; spiniform short,
about 0.25 of clasper. Aedeagus subquadrate, longer than proctiger; dorsal sclerotiza-
tion emarginate, producing lateral apical sclerotized lobes.

PUPA (fig.54). Abdomen: 4.12 mm. Trumpet: 0.63 mm; index about 5.3. Paddle:
0.99 mm. Diagnostic characters as in the key. Chaetotaxy as given below, hairs mod-
erately to darkly pigmented. Readily distinguished from eleanorae and osornoi by all

Berlin: Subgenus Howardina 111

metanotal hairs single; from marinkellei by dorsally placed hair 8-VII. Cephalotho-
rax: Mesonotum moderately pigmented between trumpets, lateral areas and metano-
tum darker; trumpets strongly contrasting. Hair 1(1), 2(1), 3(1), 4(2, 2-3), 5(2), 6(1),
7(1, 1-2), 8(1), 91), 101), 1101), 12(1). Abdomen: Tergite I strongly pigmented ex-
cept for usual membranous areas; rest of segments with moderate to strong pigmen-
tation, progressively lighter caudad. Segment I: hair 1(9-12 primary branches, each
with many secondary ones), 2(1), 3(1), 4(1, 1-2), 5(1, 1-2), 6(1), 7(1), 9(1). Seoment
IT: hair 0(1), 1(2, 1-3), 3€1, 1-2), 4(2), 5(1), 6(1), 7(1), 8(1), 9(1). Segment ITI: hair
00), 101), 201), 301); 404), 5), 6€1), 7045 1-2),80), 91); 10014 104), 140) Sex
ment IV: hair 0(1), 11, 1-3), 2(1), 31, 1-2), 4(1), 5(2), 6(1), 71, 1-2), 8(1), 9€1),
10(1), 11(1), 141). Segment V: hair 0(1), 1(1), 2(1), 3(1), 4(1), 5(2), 6(1), 7(2),
8(1), 9(1), 10(1), 11(1), 14(1). Segment VI: hair 0(1), 1(2), 3(1), 4(1), 5(1), 6(1),
71), 8C1), 9C1, 1-2), 1001), 111), 14€1). Segment VII: hair 0(1),:1(1; 1-2), 2(1),
3(1), 4(1), S(1), 6(2, 1-2), 7(1), 8(1), 9(4, 2-4), 10(1), 11(1), 14(1). Segment VIII:
hair O(1), 4(1), 9(7, 6-7), 14(1). Paddle: Uniformly moderately pigmented, midrib
darker; external margin serrated; apex faintly emarginate; hair 1-P slender. Male geni-
tal lobe extending to 0.44 of paddle, female genital lobe to 0.31.

LARVA (fig.55). Head: 0.94 mm. Siphon: 1.05 mm. Anal Saddle: 0.28 mm. Diag-
nostic characters as in the key. Chaetotaxy as given below; hairs moderately to dark-
ly pigmented. Readily separated from marinkellei by hair 5-C single; from osornoi by
hair 3-X double; from eleanorae by pecten teeth terminating before hair 1-S. Head:
Width subequal to length; pigmentation strong except around mouthparts where it is
deeper. Mental plate with a strong median tooth and usually 12(12-14) distinct teeth
on either side, often a smaller detached tooth at base. Hair 0(1), 1(1), 3(1), 4(7, 4-7),
SY, T-2). 6(3, 2-3), 7(4, 4-5), 8€1, 1-2), 9(3), 10(3, 2-3), 11(16, 10-16, stellate,
branches of varying length), 12(1), 13(1), 14(15, 10-19), 15(], '-2); Antenna: Length
about 0.38 of head; pigmentation strong; shaft with a few scattered spicules. All hairs
except I, single; hair 1-A (2, 2-3) at about 0.64 from base. Thorax: Integument weak-
ly spiculate, spicules hairlike. Apices of long hairs attenuate; barbs present on long
and stellate hairs. Prothorax: hair 0(26, 26-44), 1(2, 1-2), 2(1), 3(44, 34-72), 4(1, 1-
2), 5(1), 6(1), 7(2), 8(24, 23-36), 9(24, 13-34), 10(1), 11(1, 1-3), 12(1), 13(34, 29-
51), 14(36, 36-50). Mesothorax: hair 1(20, 9-32), 2(1), 3(1), 4(1), 5(1), 6(5, 4-6),
7(1), 8(6, 3-8), 9(7, 6-8), 10(1), 11€1), 12(1), 13(18, 15-29), 14(13, 3-21). Metatho-
rax: hair 1(26, 21-44), 2(1), 3(4, 3-5), 4(18, 16-30), 5(16, 14-20), 6(1), 7(5, 3-7),
8(16, 14-20), 9(3), 10(1), 11(1, 1-2), 12(1), 13(26, 20-46). Abdomen: Integument
and stellate hairs as on thorax. Segment I: hair 1(22, 21-40), 2(27, 17-37), 3(1), 4(1),
S(14, 10-26), 6(2, 2-3), 7(1), 9(1, 1-2), 10(1), 11(20, 17-36), 13(19, 14-24). Segment
II: hair 0(1), 1(22, 19-45), 2(23, 17-46), 3(1), 4(1), 5(16, 11-26), 6(3, 2-3), 7(11, 9-
17), 8(1), 9(9, 8-20), 10(1), 111), 12(1), 13(18, 16-23). Segment III: hair 0(1), 1(22,
22-50), 2(22, 18-47), 3(1), 4(1), 5(15, 14-32), 6(1, 1-2), 7(11, 8-20), 8(1), 9(20,
16-34), 10(1), 11(1), 12(1), 13(25, 16-36), 14(1). Segment IV: hair 0(1), 1(24, 21-
49), 2(22, 20-47), 3(1), 4(1), 5(16, 13-31), 6(1, 1-2), 7(10, 9-18), 8(1), 9(22, 18-33),
10(1), 11€1), 12(1), 13(25, 18-40), 14(1). Segment V: hair 0(1), 1(20, 18-46), 2(22,
18-45), 3(1), 41), 5(18, 14-26), 6(1), 7(10, 8-19), 8(1), 9(18, 16-31), 10(1), 11(1),
12(1), 13(24, 17-41), 14(1). Segment VI: hair 0(1), 1(24, 22-40), 2(22, 20- 40), 3(1),
4(1), 5(16, 11-24), 6(1), 7(8, 8-16), 8(1), 9(20, 18-31), 10(1), 11(1), 12(1), 13(16,
12-16), 14(1). Segment VII: hair 0(1), 1(24, 18-27), 2(18, 16-27), 3(1), 4(1), 5(16,
12-19), 6(8, 8-11), 71), 8(10, 10-14), 9(17, 15-21), 10(1), 11(1), 12(1), 13(15, 13-
13), 14(1). Segment VIII: Comb scales in a single row, usually 14(12-15) in number;
individual scale long, pointed apically, fringed nearly to tip. Hair 0(1), 1(24, 19-24),

112 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

2(1), 3(7, 6-8), 4(1), 5(19, 11-21), 14(1). Siphon: Index about 3.8-4.3; integument
imbricate, strongly pigmented, with darker basal ring. Pecten extending to about 0.5,
usually 23(22-24) teeth; individual tooth with basal denticles. Hair 1(3, 2-3), 2(1),
3-5 not visible, 6(1), 7(1), 8(1), 9(1). Anal Segment: Saddle strongly pigmented,
with darker basal ring; caudal margin with spines of varying sizes. Gills subequai, fin-
gerlike, about 2.0 of saddle length. Hair 1(3, 2-4), 2(4, 3-5), 3(2), 4a(2), 4b(2), 4c(2),
4d(2), 4e(2, 1-2). |

SYSTEMATICS. A. leei appears to be closely related to osornoi and marinkellei.
It is sharply differentiated ecologically from the other 2 species, occurring apparent-
ly at considerably higher elevations in the paramo. All the stages are known from
the type locality of Paramo de Purace in southern Colombia at elevations of 9600-
10,500 ft (COL 16,17,19,52,53,54). The only other definite record is a single female
from Zongo near La Paz, Bolivia. It seems very probable that this species will even-
tually be found in the intermediate area in Ecuador and Peru.

BIONOMICS. The immature stages have been collected in epiphytic and terres-
trial bromeliads in the Paramo de Purace where a number of females were also taken
biting-landing and entering a car between 1715 and 1815 hrs.

DISTRIBUTION (fig.6). Paramo in southern Colombia and Bolivia, above 9000
ft. Material examined: 188 specimens; 1 6, 34 9, 31 pupae, 122 larvae; 6 individual
rearings (2 pupal, 4 incomplete).

BOLIVIA. Cochabamba: Chapare, Yungas del Palmar, elev. 3200 m, 1 ?(Prosen, Carcavallo and
Martinez 1964:104). La Paz: Zongo, 1 Apr 1946, W.H.W. Komp, 1 ? [USNM].

COLOMBIA. Cauca: Paramo de Purace, 9600-10,500 ft, 18 Feb 1965, V.H. Lee (COL 52), 1 p?
(52-101), 4 IP (52-18,19,29,36), 8 P, 38 L; 19 Feb 1965, V.H. Lee, 7 L (COL 53), 5 P, 36 L(COL
54) [UCLA]. Huila: 10.0-17.5 mi S.E. of Purace, 9600-10,500 ft, 28 Oct 1964, V.H. Lee, 1 pd
(16-100), 33 2 (COL 17, type series); 30 Oct 1964, V.H. Lee, 9 P, 25L,11(COL 19),3P, 11 L
(COL 20) [USNM, BM, UCLA].

30. Aedes (Howardina) osornoi Berlin, n.sp.
Figs.6,9 56,57

TYPES: Holotype ¢ (COB 37-16) with associated pupal and larval skins and genitalia slide,
“Chorro de Padilla,’ Rio San Francisco, Bogota, Cundinamarca, Colombia, elev. 8200 ft, 14 July
1965, J.A. Ochoa [USNM]. Allotype 2 (COB 37-10) with associated pupal and larval skins, same
data as holotype [USNM]. Paratypes: 2 6 (COB 37-11,12), 2 2 (37-17,18), with associated pupal
and larval skins, same data as holotype [BM, UCLA]. This species is dedicated to Dr. E. Osorno-
Mesa, in recognition of his important contributions to the knowledge of neotropical mosquitoes.

Aedes (Howardina) quadrivittatus in part of Dyar (1928:229); Edwards (1932:155); Lane (1939:
100; 1953:713); Reyes (1955:74,89); Stone, Knight and Starcke (1959:174); Cova Garcia, Su-
til and Rausseo (1966a:58,283; 1966b:12,141,171,334).

FEMALE. Wing: 3.8 mm. Proboscis: 3.24 mm. Forefemur: 2.34 mm. Abdomen:
about 3.8 mm. Similar to eleanorae but generally larger and differing mainly in the
following. Head: Light scales on median stripe and short orbital line narrow, pale
golden; light scales on sides and venter creamy. Erect scales predominantly pale gold-
en, a few on sides darker. Palpus about 0.15 of proboscis; predominantly dark, with
white scales on apex and mesal surface of segment 4. Mesal scale patch on torus
creamy. Thorax (fig.9): All mesonotal light lines golden; outer dorsocentral line
joined anteriorly to entire lateral prescutal area in fossa, gradually narrowed caudad

Berlin: Subgenus Howardina 113

to scutellum; lateral marginal line separate only from scutal angle to wingbase; scales
on anterior promontory golden. Light scales on median scutellar lobe predominantly
narrow, pale golden, a few basal ones broader; lateral lobe with a few narrow dark
and creamy scales. Scales on paratergite golden, moderately broad. Apn scales pre-
dominantly broad, pale golden, but a few anterior scales narrower; ppn only with an
upper patch of narrow golden scales; psp scales narrow, pale golden; other patches
pure white and distributed as in eleanorae. Legs: Forefemur predominantly dark an-
teriorly except for a few basal creamy scales, posterior surface mostly dark except
for creamy longitudinal line from base to apex; midfemur dark anteriorly except for
a few basal creamy scales and a dingy white spot at about 0.5, posterior surface
creamy in basal 0.5 and along a narrow distal line, rest dark; hindfemur creamy in
basal half. Foretarsus with white scales on segments 1,2; midtarsus with short incom-
plete white rings on 1,2, occasionally a few white scales on 3; hindtarsus with broad-
er white rings on 1-3. Haltere: Stem pale, knob predominantly creamy, a few dorsal
scales dark. Abdomen: Basomedian tergal light scales absent. Sternites with broad
basal V-shaped silvery white patches; apical white scales inconspicuous.

MALE. Wing: 4.5 mm. Proboscis: 3.24 mm. Forefemur: 2.16 mm. Essentially sim-
ilar to female in color. Palpus entirely dark.

MALE GENITALIA (fig.56). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with 5-7 strong setae. Sidepiece length 2.5-3.0 of median
width. Clasper about 0.67 of sidepiece, with 2 subapical hairs; spiniform moderate,
length about 0.33 of clasper. Aedeagus subquadrate, shorter than proctiger; dorsal
sclerotization shallowly emarginate, lateral lobes inconspicuous.

PUPA (fig.56). Abdomen: 4.29 mm. Trumpet: 0.61 mm; index 6.3-6.4. Paddle:
0.94 mm. Chaetotaxy as given below, hairs moderately to darkly pigmented. Distin-
guishable from Jeei and marinkellei by hairs 10,12-C branched; from eleanorae by the
ventrally placed hair 8-VII. Cephalothorax: Mesonotum lightly pigmented between
trumpets; lateral areas and metanotum slightly darker; trumpets strongly contrasting.
Hair 1(1), 2(1), 31), 4(3, 1-4), 5@, 1-3), 6(1, 1-2), 7(2, 1-3), 8(2, 1-3), 9(1), 103,
1-4), 11(1), 12(@, 1-4). Abdomen: Tergite I moderately pigmented except for usual
membranous areas; tergites II,III also moderately pigmented, distal segments pro-
gressively lighter caudad. Segment I: hair 1(12-21 primary branches, subequal in
length and each with a few branches), 2(1, 1-3), 3(1), 4(2, 2-3), 5(2, 1-3), 6(1), 7(1),
9(1), 111, 1-2 only occasionally present). Segment II: hair 0(1), 1(9, 3-19), 2(1, 1-
2), 3(1), 4(2, 1-3), 5C1, 1-2), 6(1), 7(1, 1-2), 8(1, 1-2), 9(1). Segment III: hair 0(1),
as 2); 2171-3), SCl, 1-2); ACL 1-2), S01 42), CET): FO; 3-2) Be. 22 Os.
101), 1 1(1), 14G). Segment PV: hair 01); 12, 1-2); 2¢1, 1-2); 3(3; 2-3), 402); 50,
2-3), 6(1), 7(2, 1-2), 8C1, 1-2), 9(1), 10(1), 11(1), 14(1). Segment V: hair 0(1), 1(1),
211-3); 3€1), 40, 1-2), 5@2; 1-2), 6(1), 70; 1-3), 8@, 1-2); 901), 1001), 1101),
14(1). Segment VI: hair 0(1), 1(1), 21, 1-2), 3(1), 4€1, 1-2), 5(1, 1-2), 6(1), 7d, 1-
2), 8(2, 1-3), 9(1), 10(1), 111), 14(1). Segment VII: hair 0(1), 1(1), 2(1), 3(1), 4(1),
S(1), 6(2, 1-2), 7(1), 8@, 1-3), 9(3, 2-4), 10(1), 11€1), 14(1). Segment VIII: hair 0(1),
4(1, 1-3), 9(8, 5-9), 14(1, 1-2). Paddle: Uniformly lightly pigmented, midrib darker;
external margin serrated and apex rounded; hair 1-P slender. Male genital lobe ex-
tending to about 0.47 of paddle, female genital lobe to 0.3.

LARVA (fig.57). Head: 0.99 mm. Siphon: 0.91 mm. Anal Saddle: 0.33 mm. Read-
ily separated from other members of the group by hair 3-X single. Chaetotaxy as
given below. Head: Width subequal to length; pigmentation moderate except around
mouthparts where it is deeper. Mental plate with a strong median tooth and 11 dis-
tinct teeth on either side, often a smaller detached tooth at base. Hair 0(1), 1(1),

114 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

3(1), 4(6, 4-8), 5(1), 6(3, 3-4), 7(4, 4-5), 8(1), 9(2, 1-2), 10(1), 11°16, 9-16), 12(1),
13(1), 14(12, 7-16), 15(1, 1-2). Antenna: Length about 0.33 of head; pigmentation
light, shaft with a few scattered spicules. All hairs except 1 single; hair 1-A (2, 2-3)
about 0.64 from base. Thorax: Integument glabrous. Apices of long hairs attenuate;
stellate hairs with unequal, weakly barbed, spikelike branches. Prothorax: hair 0(28,
24-33), 1(2),. 2011-2), 3G1,.27:36),.4(01), SC), 601), 7(2), 826, 20-29), 9016, 15-
21), 10(1), 11(2, 2-3), 12(1), 13(30, 29-37), 14(41, 38-45). Mesothorax: hair 1(22,
18-25), 2(1), 3(1), 4(1), 51), 6(4, 3-4), 7(1), 805, 4-6), 9(5, 4-7), 10(1), 11€1, 1-2),
12(1), 13(18, 13-20), 14(13, 11-15). Metathorax: hair 1(25, 23-37), 2(1), 3(16, 12-
16), 4(3, 3-8), 5(16, 14-19), 6(1), 7(4, 3-4), 8(16, 15-16), 9(3, 3-4), 10(1), 11€1, 1-2),
12(1), 13(20, 18-23). Abdomen: Integument weakly spiculate, spicules simple, hair-
like; stellate hairs as on thorax. Segment I: hair 1(18, 18-21), 2(17, 16-22), 3(1),
4(1), 5(16, 13-18), 6(3, 2-3), 7(1), 9(6, 46), 10(1), 11016, 15-18), 13(16, 15-18).
Segment II: hair 0(1), 1(20, 17-22), 2(20, 16-20), 3(1), 4(1), 5(16, 14-18), 6(2, 2-3),
7(12, 10-12), 8(1), 9(6, 6-9), 10(1), 11(1), 121), 13(24, 21-24). Segment III: hair
O(1), 1(18; 18-22), 2017, 17-20), 3(1), 4(1), 5(16, 13-20), 6(1), 7(11, 10-13), 81),
9(14, 11-14), 10(1), 11(1), 121), 13(23, 21-28), 14(1). Segment IV: hair 0(1), 1(21,
19-24), 2(20, 16-22), 3(1), 4(1), 5(17, 16-19), 61, 1-2), 7€11, 10-13), 8(1), 9C15,
14-18), 10(1), 11(1), 12(1), 13(22, 18-25), 14(1). Segment V: hair 0(1), 1(21, 19-
22), 2(20, 17-22), 3(1), 41), 5(17, 16-19), 6(2, 1-2), 7010, 10-14), 8(1), 917, 14-
20), 10(1), 11(1), 12(1), 13(22, 20-25), 14(1). Segment VI: hair 0(1), 1(21, 19-21),
-2(22, 17-24), 3(1), 4(1), 5(16, 14-18), 6(1), 7(12, 10-14), 8(1), 9(16, 14-19), 10(1),
11(1), 12(1), 13(24, 22-26), 14(1). Segment VII: hair 0(1) 1(22, 20-24), 2@20, 18-
22), 3(1, 1-2), 41), 5(18, 17-21), 6(9, 9-11), 71), 8(17, 14-19), 9(16, 14-17), 10Q1),
11(1), 12(1), 13(24, 20-27), 14(1). Segment VIII: Comb scales in a single row, usual-
ly 16(14-16) in number; individual scale long, pointed apically, fringed nearly to tip.
Hair 0(1), 1(24, 21-26), 2(1), 3(6, 5-7), 4(1), 5(24, 18-24), 14C1, 1-2). Siphon: Index
about 3.3-3.67; integument moderately pigmented, with darker basal ring. Pecten
extending to about 0.42-0.5, usually 18(16-22) teeth; individual tooth with basal
denticles. Hair 1(4, 3-4), 2(1), 3-5 not visible, 6(1), 7(1), 8@2, 1-4), 9(1). Anal Seg-
ment: Saddle moderately pigmented, with a darker basal ring; caudal margin with
spines of varying sizes. Gills subequal, more than 2.0 of saddle length. Hair 1(2), 2(4,
3-4), 3(1), 4a(2, 2-3), 4b(2), 4c(2), 4d(2, 1-2), 4e(1, 1-2).

SYSTEMATICS. A. osornoi is readily separated from its apparent nearest rela-
tives, marinkellei and leei, by constant diagnostic features in all known stages. Imma-
ture stages and females are known only from elevations above 8,000 feet in Cundina-
marca Province, Colombia, in the central part of the Cordillera Oriental. A female
collected at 9,000 feet in the Sierra Nevada of the Cordillera de Merida in Venezuela
[USNM] is undoubtedly osornoi and it seems very likely that the record of quadri-
vittatus from Aragua in the Cordillera de la Costa (Cova Garcia, Sutil and Rausseo
1966) also pertains to this species. In the fair sample of material no striking variation
is apparent except for 1 reared female from Cundinamarca Province (COB 37-13)
where osornoi is sympatric with marinkellei. In this specimen a suggestion of possi-
ble hybridization or introgression of these 2 species is indicated in the osornoi-like
ornamentation of the adult and in the resemblance of the immature stages to marin-
kellei. In this area of sympatry, the 2 species have not been collected together in the
same bromeliad axils, suggesting an ecological barrier of some kind between them.

BIONOMICS. The known breeding sites of osornoi are epiphytic bromeliads at
elevations above 8,000 feet. As in the case of marinkellei, several females have been
taken, probably in biting-landing collections, at human bait.

Berlin: Subgenus Howardina 115

DISTRIBUTION (fig.6). Colombia (Cordillera Oriental) and Venezuela (Cordillera
de Merida and probably Cordillera de la Costa) at elevations of 8000-9000 ft. Materi-
al examined: 114 specimens; 11 6, 45 9, 22 pupae, 36 larvae; 22 individual rearings
(20 larval, 2 incomplete).

COLOMBIA. Cundinamarca: Bogota, B.G. Amortegui, 5 ? [USNM]. Puerto Lievano and Bogo-
ta, 6 Aug 1935, B.G. Amortegui, 9 9 [USNM]. Rio San Francisco, 21 May 1942, W.H.W. Komp,
7 2 [USNM]; “Chorro de Padilla,” elev. 8000 ft, 9 Feb 1965, E. Ruiz, 1 lpd (11-10), 1 lp? (11-11),
3 1(COB 11), 12 2, 10 L(COB 650209); 25 Feb 1965, E. Ruiz, 21(COB 14); 21 June 1965, J.A.
Ochoa (COB 34), 3 Ipd (34-18,93,94), 5 Ip? (34-14-17,19), 2 lp (34-95,98), 1 3, 1 9; 14 July
1965, J.A. Ochoa (COB 36), 2 Ipé (36-21,22), 1 Ip? (36-20), (COB 37, type series), 3 Ipd (37-11,
12,16), 3 lp? (37-10,17,18) [USNM, BM, UCLA].

VENEZUELA. Aragua: (Cova Garcia. Sutil and Rausseo 1966b:141). Merida: Sierra Nevada,
elev. 9000 ft, Dec 1923, E.P. de Bellard, 1 9? [USNM].

31. Aedes (Howardina) ecuadoriensis Berlin, n.sp.
Figs.6,9,58

TYPES: Holotype ? (ECU 4), Leon River, Saraguro, Azuay, Ecuador, 5500 ft, adult collection,
22 Mar 1965, L.E. Pena G. [USNM]. Paratypes: 4 9 (ECU 4), same data as holotype [UCLA].

FEMALE. Wing: 2.88 mm. Proboscis: 2.25 mm. Forefemur: 1.62 mm. Abdomen:
about 2.25 mm. Similar to eleanorae, with the following diagnostic differences. Head:
Light scales on median and short orbital lines moderately broad, white; sides and
venter with white scales. Erect scales dark. Palpus about 0.15 of proboscis; predomi-
nantly dark, with white scales on apex and base of segment 4. Mesal patch on torus
white. Thorax (fig.9): Mesonotal light lines white and golden; inner dorsocentral
line complete, narrow, white; outer dorsocentral line apparently complete, narrow,
pale golden, extending from anterior margin to scutellum, a few anterior scales whit-
ish; lateral marginal line short, whitish, extending from scutal angle to wingbase;
scales on anterior promontory white. Light scales on median scutellar lobe broad,
white, narrow and brown on lateral lobes. Scales on paratergite broad and white. Apn
scales moderately broad, white; ppn with large upper and small lower scale patch;
former with narrow white scales above, broad below; latter with broad white scales;
other scale patches broad, white scaled, distributed as in the group. Legs: Forefemur
dark scaled anteriorly, posterior surface predominantly creamy except for narrow
longitudinal dark line from base to apex; midfemur and hindfemur mostly dark ante-
riorly except for a ventral white line to middle and a white spot at 0.67, posteriorly
proximal 0.5 creamy, rest dark. Foretarsus with short incomplete white rings on seg-
ments 1,2; midtarsus with short white rings on 1-3; hindtarsus with broader white
rings on 1-4. Haltere: Stem pale, knob predominantly white, a few brown scales dor-
sally. Abdomen: Basomedian tergal light scales dingy white on II-VII. Sternites with
broad basal V-shaped patches of white scales; apical light scales distinct up to seg-
ment V.

MALE. Wing: 3.06 mm. Proboscis: 2.71 mm. Forefemur: 1.75 mm. Essentially
similar to female in color. Palpus predominantly dark, basoventral parts of segments
4 and 5 white scaled.

MALE GENITALIA (fig.58). Diagnostic characters as in the key. Ninth tergite
lobe prominent, with 6,7 moderate to strong bristles, inner markedly flattened and
almost leaflike. Sidepiece length about 3.5 of median width. Clasper about 0.5 of
sidepiece, with 2,3 subapical hairs; minute spicules in basal 0.5; spiniform short, 0.25

116 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

of clasper. Aedeagus short; dorsal sclerotization deeply emarginate, producing dis-
tinct lateral sclerotized lobes.

PUPA, LARVA unknown.

SYSTEMATICS. A. ecuadoriensis is readily separated from the apparently closely
related brevivittatus by constant differences in mesonotal, pleural and tarsal mark-
ings. It is known only from 2 widely spearated localities in northern and southern
Ecuador in semi-arid areas (chaparral and high desert) at elevations between 5,000
and 6,000 feet, whereas brevivittatus is known at present only from intermediate lo-
calities with a range of elevations from 4,200 to nearly 11,000 feet, apparently al-
ways in humid forests.

BIONOMICS. Most of the known material of ecuadoriensis consists of females
taken in biting-landing collections between the hours of 0600 and 1500. The imma-
ture stages will probably be found in bromeliads which undoubtedly contain water
during the period of seasonal rainfall characteristic of the areas where the adults were
collected.

DISTRIBUTION (fig.6). Ecuador (chaparral and high desert areas) at elevations
of 5000-6000 ft. Material examined: 10 specimens; 2 6, 8 9; no individual rearings
or immature stages.

ECUADOR. Azuay: Saraguro, Leon River, elev. 5500 ft, 22 Mar 1965, L.E. Pena G., 5 9? (ECU
4), type series [USNM, UCLA]. Jmbabura: (barra, Taguando River, elev. 5400-5800 ft, 9 June

1965, LE. PenaG., 2 6, 3 9 (ECU 25) [UCLA].

32. Aedes (Howardina) brevivittatus Berlin, n.sp.
Figs.6,9,58

TYPES: Holotype ¢, Zorrocucho, Azuay, Ecuador, 3500 m, date not specified, Levi-Castillo
[USNM]. Allotype ¢ with associated genitalia slide, same data as holotype [USNM]. Paratypes:
8 2, same data as holotype [USNM, UCLA].

Aedes (Howardina) quadrivittatus in part of Dyar (1928:229); Edwards (1932:155); Lane (1939:
100; 1953:713); Levi-Castillo (1953:40); Stone, Knight and Starcke (1959:174).

FEMALE. Wing: 5.04 mm. Proboscis: 3.27 mm. Forefemur: 2.79 mm. Abdomen:
about 4.55 mm. Similar to eleanorae, differing in larger size and in the following.
Head: Light scales of median and short orbital lines creamy; light scales on sides and
venter whitish. Erect scales predominantly brown, a few middle ones paler. Palpus
about 0.14 of proboscis; predominantly dark, with white scales on apex and mesal
side of segment 4. Mesal patch on torus white. Thorax (fig.9): All mesonotal light
lines creamy to pale golden; inner dorsocentral line slightly broadened on anterior
0.5, gradually narrowing caudad; outer dorsocentral line shorter, from anterior 0.25
to scutellum; lateral marginal line longer, from anterior promontory to wingbase;
scales on anterior promontory golden. Light scales on median scutellar lobe narrow,
white; lateral lobes with narrow, brown scales. Scales on paratergite broad, dingy
white. Apn scales broad, dingy white, a few anterior ones narrower; ppn with only an
upper patch of predominantly narrow, creamy scales, a few broader ones below; oth-
er pleural patches with broad white scales, distributed as in the group. Legs: Forefe-
mur dark brown anteriorly, posteriorly proximal 0.5 predominantly creamy extend-
ing distally as a narrow creamy line, rest dark; midfemur mostly dark brown anteri-
orly except for a few basal creamy scales and a white spot at 0.67, posteriorly prox-

Berlin: Subgenus Howardina 117

imal 0.5 creamy, extending distally as a creamy line, rest dark; hindfemur creamy to
0.75 anteriorly, 0.67 posteriorly, rest dark scaled. Foretarsus with a few white scales
on segments 1,2; midtarsus with short incomplete white rings on 1-3; hindtarsus with
short white rings on 1-4. Haltere: Stem pale, knob predominantly dark, fringe scales
whitish. Abdomen: Basal tergal light bands creamy to V, white on VI,VII. Sternites
with broad basal patches, creamy basally and white on sides; apical creamy to white
scales distinct up to segment V.

MALE. Wing: 5.6 mm. Proboscis: 3.45 mm. Forefemur: 2.37 mm. Essentially sim-
ilar to female in color. Palpus predominantly dark, basoventral parts of segments 4
and 5 white scaled.

MALE GENITALIA (fig.58). As figured; diagnostic characters as in the key. Ninth
tergite lobe prominent, with 6,7 strong bristles. Sidepiece length about 2.5 of median
width. Clasper about 0.5 of sidepiece, with 1,2 subapical hairs; spiniform moderate,
about 0.33 of clasper. Aedeagus subquadrate; dorsal sclerotization deeply emar-
ginate, producing distinct lateral lobes.

PUPA, LARVA unknown.

SYSTEMATICS. A. brevivittatus is apparently closely related to ecuadoriensis
with which it is broadly sympatric in Ecuador but from which it seems to be isolated
by a strong ecological barrier. While ecuadoriensis is known only from semi-arid lo-
calities at elevations of 5,000 to 6,000 feet, brevivittatus has been collected only in
humid forests at elevations from 4,200 to nearly 11,000 feet, apparently only on the
western slopes of the Andes.

BIONOMICS. The immature stages of brevivittatus will undoubtedly be found in
epiphytic and/or terrestrial bromeliads. The present collections consist mostly of fe-
males taken biting-landing on humans between the hours of 1000 and 1800.

DISTRIBUTION (fig.6). Humid forests on western slopes of Andes in Ecuador at
elevations of 4200-11,000 ft. Material examined: 25 specimens; 8 6, 17 9; no indi-
vidual rearings or immature stages.

ECUADOR. Azuay: Cuenca, Caserio Tinajillas, elev. 10,000 ft, 18-21 Mar 1965, L.E. PenaG.,
7 3, 7 2 (ECU 3) [UCLA]. Zorrocucho, elev. 3500 m, Levi-Castillo, 1 6,9 9, type series [USNM,
UCLA]. Canar: (Levi-Castillo 1953:40). Pichincha: Santo Domingo, Tandapi, elev. 4400-5400 ft,
20-21 June 1965, L.E. Pena G., 1 9? (ECU 27) [UCLA].

Aurivittatus Group

FEMALES. Head: Median light line on vertex with narrow scales. Proboscis dis-
tinctly longer than forefemur. Torus with restricted mesal scale patch. Thorax: Meso-
notal light lines usually creamy to golden; inner dorsocentral line complete; outer
dorsocentral line either complete or not reaching anterior margin, subparallel to in-
ner; lateral marginal line continuous; scales on anterior promontory pale golden. An-
tealar area above paratergite with golden scales. Paratergite with a moderate scale
patch extending entire sclerite length. Upper ppn entirely covered with light scales;
upper stp patch quadrate. Legs: Distal white spot on anterior surface of midfemur
distinct, absent on hindfemur. Tibiae completely dark. Wing: Remigium with white
scales. Abdomen: Basomedian tergal light markings absent.

MALES. Palpus with basomedian light scales on segments 4 and 5.

MALE GENITALIA. Lobe of tergite [X distinct, with 5,6 strong bristles. Side-
pieces separated at base by a distance less than width of aedeagus; without any indi-
cation of lobes on tergomesal or sternomesal margins. Claspette represented by a

118 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

conical tubercle. Aedeagus shorter than proctiger; apex slightly een pro-
ducing lateral lobes.

PUPAE and LARVAE unknown.

DISCUSSION. The Aurivittatus Group is recognized here for the nominate form,
vanemdeni and martinezi on the basis of the presence of pale scales on the remigium.
The immature stages are unknown for the group and I have not seen any males. The
group is doubtfully distinct from the Eleanorae Group and may have to be merged
with it if the immature stages prove to be similar. Members of this group have been
occasionally misidentified as sexlineatus in the past.

The group occupies the southernmost portion of the distribution of the section.
A. aurivittatus and vanemdeni have been found only in Bolivia, at moderate eleva-
tions in the Amazon basin; martinezi is known at moderate elevations in the Amazon
basin in Bolivia and from northern Argentina at an elevation of about 5,000 feet and
is one of the 2 members of the section reported from the La Plata drainage (fig.5).

33. Aedes (Howardina) aurivittatus Cerqueira
Figs.5 ,9

1943.Aedes (Howardina) aurivittatus Cerqueira, 1943a:7-8. TYPE: Holotype ?, Pampagrande,
Florida, Santa Cruz, Bolivia, Feb 1940 [IOC].

Aedes (Howardina) aurivittatus of Cerqueira (1943b:33); Del Ponte, Castro and Garcia (195 1:240);
Lane (1953:711); Stone, Knight and Starcke (1959:173); in part of Martinez, Prosen and Car-
cavallo (1964:53-58); Prosen, Carcavallo and Martinez (1964:103); Forattini (1965:461); Stone
(1967:208).

FEMALE. Wing: 3.24 mm. Proboscis: 2.16 mm. Forefemur: 1.89 mm. Abdomen:
about 2.88 mm. Head: Light decumbent scales on head white. Erect scales extending
to vertex, light brown on occiput. Palpus about 0.16 of proboscis; white scales re-
stricted to base and apex of segment 4. Mesal scale patch on torus silvery white. Tho-
rax (fig.9): Mesonotal light lines creamy to pale golden; inner dorsocentral line com-
plete, extending from anterior margin to scutellum; outer dorsocentral line complete,
merging with lateral marginal line at level of humeral bristles; lateral marginal line
long, extending from humeral bristles to wingbase. Median scutellar lobe with broad
dark scales laterally and moderately broad creamy scales from base to apex in mid-
dle; lateral lobe with a few narrow creamy scales. Paratergite with moderately broad
white scales. Scales on apn predominantly broad and white, a few anterior ones nar-
rower; ppn with large upper and small lower patches, former predominantly with
narrow scales except for lower broader ones, latter with broad white scales; psp with
a few broad white scales restricted to posterior part of bristles; other patches pure
white, distributed as in the group. Legs: Forefemur mostly dark except for a few
white scales at base anteriorly and a creamy longitudinal line posteriorly; midfemur
predominantly dark anteriorly except for a narrow mesal line from base terminating
in a distinct silvery spot at 0.67, posteriorly proximal 0.5 creamy, rest dark; hindfe-
mur mostly creamy to 0.67 on both sides, rest dark scaled. Foretarsus and midtarsus
with a few white scales on segment 1; hindtarsus with moderately broad white rings
on segments 1-3. Haltere: Stem pale, knob brown scaled basally, fringe scales whit-
ish. Abdomen: Sternites with broad V-shaped basal white patches; median apical
white scales few, present up to segment V.

Berlin: Subgenus Howardina 119

MALE. Not seen; according to the description of Martinez, Prosen and Carcavallo
(1964:53-58), essentially similar to female as described above.

MALE GENITALIA. Diagnostic characters as in the key. Ninth tergite lobe prom-

inent, with 6 strong bristles. Sidepiece length about 3.0 median width. Clasper about
0.5 of sidepiece, with 1 subapical hair; spiniform long, about 0.5 of clasper. Aedeagus
shorter than proctiger; dorsal sclerotization slightly emarginate, producing lateral
lobes. |

PUPA, LARVA unknown.

SYSTEMATICS. The 3 females of aurivittatus | have examined are readily sepa-
rated from martinezi by the diagnostic characters mentioned above. This species ap-
pears to be restricted to moderate elevations in the upper parts of the Amazon drain-
age in Bolivia. The record of aurivittatus from La Caldera (Salta), Argentina, is prob-
ably erroneous and based on the closely related martinezi which has been collected
in the same locality. :

BIONOMICS. The immature stages are unknown but should be found in the axils
of epiphytic and/or terrestrial bromeliads, the only known breeding sites of the mem-
bers of the Sexlineatus Section. Females of aurivittatus are evidently at least occa-
sionally attracted to humans, presumably in the daytime.

DISTRIBUTION (fig.5). Bolivia (upper Amazon drainage) at moderate elevations.
Material examined: 3 specimens; 3 9; no individual rearings or immature stages.

BOLIVIA. Cochabamba: Chapare, 27 Apr 1944, 1 2 [USNM]. Santa Cruz: Florida, Feb 1940,
N.L. Cerqueira, 2 ? paratypes [USNM].

34. Aedes (Howardina) martinezi Berlin, n.sp.
Figs.5 ,9

TYPE: Holotype ? (ARG-1), La Caldera, Salta, Argentina, 27 Mar 1966, A. Martinez [USNM].
Paratypes: 30 2, same data as holotype [USNM, BM, UCLA]. This species is dedicated to Dr. A.
Martinez for his contributions to the study of neotropical mosquitoes.

Aedes (Howardina) aurivittatus in part of Martinez and Prosen (1955:25-27); Castro, Garcia and
Bressanello (1959:550).

FEMALE. Wing: 3.42 mm. Proboscis: 2.36 mm. Forefemur: 2.19 mm. Abdomen:
2.91 mm. Essentially similar to aurivittatus, differing apparently only in the follow-
ing. Head: Light decumbent scales dull creamy. Erect scales on occiput largely
brownish, a few darker. Thorax (fig.9): Mesonotal light lines pale golden; outer dor-
socentral line shorter, arising at anterior 0.25; lateral marginal line long, from side of
anterior promontory to wingbase. Lateral scutellar lobes with a few narrow dark
scales.

MALE, PUPA, LARVA unknown.

SYSTEMATICS. A. martinezi has been confused in the past with aurivittatus
from which it is clearly differentiated by the diagnostic characters mentioned above
which are remarkably constant in the small type series. Apparently martinezi extends
from moderate elevations in the Amazon basin in Bolivia to the upper part of the La
Plata drainage in northern Argentina.

BIONOMICS. The immature stages are unknown but will probably be found in
the axils of bromeliads. The adults have been captured biting and landing on humans
during the day at elevations of about 5,000 feet.

120 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

DISTRIBUTION (fig.5). Bolivia (upper Amazon drainage) and Argentina (upper
La Plata drainage), at moderate to high elevations. Material examined: 32 specimens;
32 2, no individual rearings or immature stages.

ARGENTINA. Salta: La Caldera, 3.0 mi before mountain pass Sta. Laura, elev. 5000 ft, 27
Mar 1966, A. Martinez, 31 9 (ARG-1, type series) [USNM, BM, UCLA].

BOLIVIA. Cochabamba: Chimore, 30 Apr 1944,1 9 [USNM].

35. Aedes (Howardina) vanemdeni Martini
Figs.5 ,9

1931.Aedes vanemdeni Martini, 1931:208-209. TYPE: Lectotype 6, Yungas de Coroico (La Paz),
Bolivia, elev. 1000 m [BM; selection of Mattingly 1955:31].

Aedes (Howardina) vanemdeni of Lane (1939:101); Martinez (1950:39); De] Ponte, Castro and
Garcia (1951:240); Lane (1953:716); Martinez and Prosen (1953:28,30; 1955:32); Castro, Gar-
cia and Bressanello (1959:550); Stone, Knight and Starcke (1959: 174): Prosen, Carcavallo and
Martinez (1964:105).

FEMALE. Not seen.

MALE. Essentially similar to aurivittatus, differing apparently only in the follow-
ing (fig.9). Lateral marginal line short, not connected to outer dorsocentral line at
level of humeral bristles and extending from scutal angle to wingbase. Forefemur and
midfemur with a distinct white spot anteriorly at distal 0.67. Foretarsus and midtar-
sus with white scales on segments 1-3; hindtarsus with white ring only on segment |;
rest dark scaled. Haltere: Stem pale, knob entirely white scaled.

MALE GENITALIA. Ninth tergite lobe distinct, with 5 strong bristles. Sidepiece
length about 3.0 of median width. Clasper and spiniform as in aurivittatus.

PUPA and LARVA unknown.

SYSTEMATICS. A. vanemdeni is readily differentiated from other members of
the Aurivittatus Group by the distribution of the tarsal white markings. The male
genitalia appear to be similar to aurivittatus. A. vanemdeni is known only by the 6
(lectotype) and 2 of the original type series from moderate elevations in the Amazon
basin in Bolivia.

BIONOMICS. The unknown immature stages will probably be found in bromeliad
axils. Females are presumably at least occasionally attracted to man.

DISTRIBUTION. Bolivia (fig.5). Material examined: d lectotype examined by
J.N. Belkin [BM].

Berlin: Subgenus Howardina 121

REFERENCES CITED

Aitken, T.H.G.
1960. A survey of Trinidadian arthropods for natural viral infections (August 1953
to December 1958). Mosquito News 20:1-10.
Anduze, P.J.
1941. Lista provisional de los zancudos hematofagos de Venezuela (Diptera:Culici-
dae). Bol. Entomol. Venezolana 1:6-18.
1943. Estudios de entomologia medica en el estado Merida-Venezuela. Bol. Ento-
mol. Venezolana 2:189-196.
1947. Contribucion al estudio de los vectores de la fiebre amarilla en Venezuela.
Acad. Cien. Fis. Mat. Natur., Caracas, Bol. 10:331-373.
Antunes, P.C.A.
1937a. Informe sobre una investigacion entomologica realizada en Colombia. Co-
lombia, Univ. Nac., Bogota. Fac. Med., Rev. 6:3-29.
1937b. Informe sobre una investigacion entomologica realizada en Colombia. Co-
lombia, Univ. Nac., Bogota. Fac. Med., Rev. 6:65-87.
Arnett, R.H.
1949. Notes on the distribution, habits, and habitats of some Panama culicines
(Diptera:Culicidae). N.Y. Entomol. Soc., J. 57:233-251.
1950. Notes on the distribution, habits, and habitats of some Panama culicines
(Diptera:Culicidae). N.Y. Entomol. Soc., J. 58:99-115.
Bates, M.
1944. Observations on the distribution of diurnal mosquitoes in a tropical forest.
Ecology 25:159-170.
1945. Observations on climate and seasonal distribution of mosquitoes in Eastern
Colombia. J. Anim. Ecol. 14:17-25.
1949. The natural history of mosquitoes. New York, MacMillan. 379 p.
Belem Virus Lab.
1964. Annual Report. Belem, Para, Brazil. 133 p.
Belkin, J.N.
1955. The Tripteroides caledonica complex of mosquitoes in Melanesia (Diptera:
Culicidae). Pacific Sci. 9:221-246.
1962. The mosquitoes of the South Pacific. v.1. Berkeley, U. Calif. Press. 608 p.
1968. Mosquito Studies (Diptera, Culicidae). IX. The type specimens of New
World mosquitoes in European museums. Amer. Entomol. Inst., Contrib. 3(4):
1-69.
Belkin, J.N., C.L. Hogue, P. Galina T.H.G. Aitken, R.X. Schick and W.A. Powder
1965. Mosquito Studies (Diptera, Culicidae). II. Methods for the collection, rear-
ing and preservation of mosquitoes. Amer. Entomol. Inst., Contrib. 1(2):19-78.
Belkin, J.N., R.X. Schick, P. Galindo and T.H.G. Aitken
1965. Mosquito Studies (Diptera, Culicidae). I. A project for a systematic study of
the mosquitoes of Middle America. Amer. Entomol. Inst., Contrib. 1(2):1-17.
Belkin, J.N., R.X. Schick and S.J. Heinemann
1965. Mociuité Studies (Diptera, Culicidae). V. Mosquitoes originally destined
from Middle America. Amer. Entomol. Inst., Contrib. 1(5):1-95.
Blanchard, R.
1905. Les moustiques. Histoire naturelle et medicale. Paris, F.R. de Rudeval.
673 p.

122 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Blanton, F.S. and E.L. Peyton
1958. Notes and distribution of Aedes mosquitoes in Panama based on a three
year light trap survey. Mosquito News 18:30-33.
Bonne, C. and J. Bonne-Wepster
1925. Mosquitoes of Surinam. Roy. Colon. Inst., Amsterdam, Afd. Trop. Hyg.
13:1-558.
Bonne-Wepster, J. and C. Bonne
1920. Diagnoses of new mosquitoes from Surinam, with a note on synonymy. In-
secutor Inscitiae Mens. 7:165-180.
Boshell-Manrique, J. and E. Osorno-Mesa
1944. Observations on the epidemiology of the jungle yellow fever in Santander
and Boyaca, Colombia, September 1941 to April 1942. Amer. J. Hyg. 40:170-
181.
Bugher, J.C., J. Boshell-Manrique, M. Roca-Garcia and E. Osorno-Mesa
1944. Epidemiology of jungle yellow fever in Eastern Colombia. Amer. J. Hyg.
39:16-51.
Carpenter, S.J., P. Galindo and H. Trapido
1952. Forest mosquito studies in an endemic yellow fever area in Panama. Mos-
quito News 12:156-164.
Carpenter, S.J. and E.L. Peyton
1952. Mosquito studies in Panama Canal Zone during 1949 and 1950 (Diptera,
Culicidae). Amer. Midland Natur. 48:673-682.
Castro, M., M. Garcia and M.D. Bressanello
1959. Diptera Culicidae Culicinae. Primeras Jornadas Entomoepidemiol. Argent.,
Buenos Aires, p.547-562.
Causey, O.R. and H.W. Kumm
1948. Dispersion of forest mosquitoes in Brazil. Preiiminary studies. Amer. J.
Trop. Med. 28:469-480.
Causey, O.R., H.W. Kumm and H.W. Laemmert
1950. Dispersion of forest mosquitoes in Brazil. Further studies. Amer. J. Trop.
Med. 30:301-312.
Causey, O.R. and G.V. Dos Santos
1949. Diurnal mosquitoes in an area of small residual forests in Brazil. Entomol.
Soc. Amer., Ann. 42:471-482.
Cerqueira, N.L.
1943a. Algumas especies novas da Bolivia, e referencia a tres especies de Haemago-
gus (Diptera, Culicidae). Inst. Oswaldo Cruz, Rio de Janeiro, Mem. 39:1-14.
1943b. Lista dos mosquitoes da Bolivia (Diptera, Culicidae). Inst. Oswaldo Cruz,
Rio de Janeiro, Mem. 39:15-36.
Coquillett, D.W.
1902. Three new species of Culex. Can. Entomol. 34:292-293.
1905. New Culicidae from West Indies and Central America. Entomol. Soc. Wash.,
Proc. 7:182-186.
1906a. Five new Culicidae from the West Indies. Can. Entomol. 38: 60-62.
1906b. A classification of the mosquitoes of North and Middle America. U.S. Bur.
Entomol., B. Tech. Ser. 11:1-31.
1910. The type species of the North American Genera of Diptera. U.S. Nat. Mus.,
Proc. 37:499-647.
Cova Garcia, P., E. Sutil and J.A. Rausseo
1966a. Mosquitos (Culicinos) de Venezuela, v.1. Caracas, Min. Sanid. Asistencia

Berlin: Subgenus Howardina 123

Soc. 410 p.
1966b. Mosquitos (Culicinos) de Venezuela, v.2._ Caracas, Min. Sanid. Asistencia
Soc. 406 p.
Davis, N.C. and R.C. Shannon
1931. Further attempts to transmit yellow fever with aS of South Ameri-
ca. Amer. J. Hyg. 14:715-722.
Del Ponte, E., M.P. Castro and M. Garcia
1951. Clava para las especies de Psorophora y Aedes de la Argentina y comorcas
vecinas. Diagnosis de Aedes (O.) raymondi n.sp. (Diptera, Culicidae). Soc. Cien.
Argent., An. 151:228-243.
Diaz Najera, A. |
1966a. Aedes de Mexico. Descripcion de una nueva especie del subgenero Howarda-
ina (Diptera, Culicidae). Rev. Invest. Salud. Publ. (Mex.) 24:331-344.
1966b. OMe tropicales de Mexico. Rev. Invest. Salud. Publ. (Mex.) 26:57-64.
Dyar, H.G.
1905a. Remarks on genitalic genera in the Culicidae. Entomol. Soc. Wash., Proc.
7:42-49.
1905b. A synoptic table of North American mosquito larvae. N.Y. Entomol. Soc.,
J,,43:22-26.
1905c. Brief notes on mosquito larvae. N.Y. Entomol. Soc., J. 13:26-29.
1906. Illustrations of mosquito larvae. Entomol. Soc. Wash., Proc. 8:15-21.
1918. The male genitalia of Aedes as indicative of natural affinities. Insecutor Ins-
citiae Mens. 6:71-86.
1920. The Classification of American Aedes. Insecutor Inscitiae Mens. 8:103-106.
1925. The mosquitoes of Panama. Insecutor Inscitiae Mens. 13:101-195.
1928. The mosquitoes of the Americas. Washington, Carnegie Inst. (Publication
387). 616 p.
Dyar, H.G. and F. Knab
1906a. Notes on some American mosquitoes with description of new species. Biol.
Soc. Wash., Proc. 19:159-172.
1906b. The larvae of Culicidae classified as independent organisms. N.Y. Entomol.
Soc., J. 14:169-230.
1907. Description of some American mosquitoes. N.Y. Entomol. Soc., J. 15:9-13.
1913. Three new Neotropical mosquitoes. Insecutor Inscitiae Mens. 1:76-78.
1919. New species of tropical American mosquitoes. Insecutor Inscitiae Mens. 7:
1-9. |
Dyar, H.G. and M. Nunez Tovar
1927. Notes on biting flies from Venezuela. Insecutor Inscitiae Mens. 14:152-154.
Edwards, F.W.
1932. Diptera fam. Culicidae. Jn P. Wytsman, Genera Insectorum, Brussells, Des-
met-Verteneuil, Fasc. 194. 258 p.
Fauran, P.
1961. Catalogue annote des Culicides signales en Guyane Francaise. Inst. Pasteur
Guyane Franc. Inini, Arch. Publication 465. 66 p.
Floch, H. and E. Abonnenc
1942a. Especes de moustiques signalees pour la premiere fois en Guyane Francaise.
Inst. Pasteur Guyane Ter. Inini, Publication 41. 6 p.
1942b. Catalogue et distribution geographique des moustiques de la Guyane Fran-
caise actuellement connus (1). Inst. Pasteur Guyane Ter. Inini, Publication 43.
10 p.

124 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

1944. Culicides et Ixodides Guyanais. Moustiques signales pour la premiere fois et
presence de Ornithodorus talaje en Guyane francaise. Inst. Pasteur Guyane
~ Ter. Inini, Publication 86. 6 p.
1945. Les Moustiques de la Guadeloupe (II). Les Genres Megarhinus, Aedes, Culex,
Deinocerites, Mansonia et Wyeomyia. Inst. Pasteur Guyane Ter. Inini, Publica-
tion 110. 48 p.
1947. Distribution des culicines des genres autres que la genre Culex, en Guyane
Francaise. Inst. Pasteur Guyane Ter. Inini, Publication 148. 12 p.
Floch, H. and P. Fauran
1954. Lutte antipaludique et lutte antiamarile en Guyane francaise. Inst. Pasteur
Guyane Franc. Inini, Arch. Publication 322. 67 p.
Forattini, O.P.
1965. Entomologia Medica. v.2. Sao Paulo, Univ. 506 p.
Galindo, P.
1964. Ecology of Arthropod-borne viruses in Panama. Gorgas Mem. Lab., Annu.
Progr. Rep. 130 p. :
Galindo, P., S.J. Carpenter and H. Trapido
193k: Ecological observations on forest mosquitoes of an endemic area in Panama.
Amer. J. Trop. Med. 31:98-137.
1955. A contribution to the ecology and biology of treehole bisects mosquitoes
of Panama. Entomol. Soc. Amer., Ann. 48:158-164.
Galindo, P. and H. Trapido
1957. Forest mosquitoes associated with sylvan yellow fever in Nicaragua. Amer.
J. Trop. Med. Hyg. 6:145-152.
Galindo, P., H. Trapido and S.J. Carpenter
1950. Observations on diurnal forest mosquitoes in relation to sylvan yellow fever
in Panama. Amer. J. Trop. Med. 30:533-574.
Gerry, B.I.
1932. Morphological studies of the female genitalia of Cuban mosquitoes. Ento-
Soc. Amer., Ann. 25:31-64.
Giles, G.M.
1902. A handbook of the Gnats or Mosquitoes. London, John Dale and Daniels-
son. 530 p.
Gowdey, C.C.
1926. Catalogus Insectorum Jamaicensis. Jamaica. Dept. Agri., Ent. Bull. 4:1-114.
Grabham, M. :
1906. Notes on some mosquitoes from Newcastle, Jamaica. Can. Entomol. 38:
167-173.
1907. Notes on some mosquitoes from Jamaica, West Indies. Can. Entomol. 39:
25-26.
Hill, R.B. and C.M. Hill
1948. The mosquitoes of Jamaica. Inst. Jamaica, B., Sci. Ser. 4. 60 p.
Horsfall, W.R.
1955. Mosquitoes, their bionomics and relation to disease. N.Y., Ronald Press.
123,
Howard, L.O., H.G. Dyar and F. Knab
1912. The mosquitoes of North and Central America and the West Indies. v.1.
Washington, Carnegie Inst. (Publication 159). 520 p.
1917. The mosquitoes of North and Central America and the West Indies. v.4.
Washington, Carnegie Inst. (Publication 159). p. 525-1064.

Berlin: Subgenus Howardina 125

Johnson, C.W.
1919. A revised list of Diptera of Jamaica. Amer. Mus. Nat. Hist., Bull. 41:421-
449.
Knight, K.L. and E.N. Marks
1952. An annotated checklist of the mosquitoes of the subgenus Finlaya, genus
Aedes. U.S. Nat. Mus., Proc. 101:513-574.
Komp, W.H.W.
1936a. An annotated list of the mosquitoes found in the vicinity of an endemic
focus of yellow fever in the Republic of Colombia. Entomol. Soc. Wash., Proc.
38:57-70.
1936b. The male and larva of Aedes dominicii Rangel & Romero Sierra, and the
male of Aedes pseudodominicii sp.nov., representatives of a new subgenus (So-
peria) of the genus Aedes, from Colombia. Entomol. Soc. Wash., Proc. 38:71-
The #
1956. Notes on mosquitoes from an area of endemic yellow fever in Colombia.
Entomol. Soc. Wash., Proc. 58:37-42.
Kumm, H.W. and W.H.W. Komp
1941. Aedes (Howardina) allotecnon, a new species of Aedes from Costa Rica, and
description of the larva, adult and male terminalia of Aedes quadrivittatus
(Coq). Entomol. Soc. Wash., Proc. 43:17-25.
Kumm, H.W., W.H.W. Komp and H. Ruiz
1940. The mosquitoes of Costa Rica. Amer. J. Trop. Med. 20:385-422.
Kumm, H.W. and O. Novis ,
1938. Mosquito studies on the Ilha de Marajo, Para, Brazil. Amer. J. Hyg. 27:498-
ee oe
Kumm, H.W. and H. Zuniga
1942. Mosquitoes of El Salvador. Amer. J. Trop. Med. 22:399-415.
van der Kuyp, E.
1954. Mosquitoes of Netherlands Antilles and their hygienic importance. Stud.
Fauna Curacao Caribbean Islands 5:37-114. _
Lane, J:
1936a. Notas sobre investigacoes entomologicas em localidades onde houve febre
amarella sylvestre em Sao Paulo. Arch. Hyg. Saude Pub., Sao Paulo 2:127-133.
1936b. Notas sobre culicideos de Matto Grosso. Mus. Paulista, Sao Paulo, Rev.
20:173-206.
1939. Catalogo dos mosquitos neotropicos. Sao Paulo, Clube Zool. Brasil. 218 p.
1949. Zoogeography of the Culicidae in the world. Mus. Paranaense, Curitiba,
Rev. 7:247-263.
1951. Synonymy of Neotropical Culicidae. Entomol. Soc. Wash., Proc. 53:333-
336.
1953. Neotropical Culicidae, v.2. Sao Paulo, Univ. p.554-1112.
Levi-Castillo, R.
1952a. Estudios epidemiologicos sobre la Fiebre Amarilla selvatica en las Provin-
cias de Manabi y esmeraldas de la zona costera de la Republica del Ecuador.
Rev. Kuba Med. Trop. 8(10,11 y 12):76-81.
1952b. Vorlaufige liste der stechmucken (Uranotaeniini, Toxorhynchitini, Culi-
cini, Aedini, and Sabethini) aus Ecuador (Diptera, Culicidae). Z. Tropenmed.
Parasitol. 3(4):552-559. :
1953. Lista provisional y distribucion de los mosquitos culicinos en el Ecuador.
Rev. Ecuator. Entomol. Parasitol. 1(1):34-45.

126 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

Lima, A. de Costa
1930. Sobre os mosquitos que se craim em buracos de arvores. Inst. Oswaldo
Cruz, Rio de Janeiro, Mem. 23:255-260.
Lutz, A.
1904. Catalogo dos culicideos Brasileiros e Sul-Americanos. /n Bourroul, C. Mos-
quitos do Brasil. 16 p.
1905. Novas especies do Mosquitos do Brasil. Imprensa Med., Sao Paulo 13:65-70.
Martinez, A.F.
1950. Algunos Culicidae nuovos o poco conocidos para las entomofaunas de Ar-
gentina, Bolivia y Paraguay. Mision Estud. Patol. Reg. Argent. 21:33-41.
Martinez, A.F., R.U. Carcavallo and A.F. Prosen
1960. Una nueva especie de Culicidae de Bolivia. Univ. Buenos Aires, Fac. Cien.
Med., Cat. Microbiol. Parasitol., Rev. 28:25-29.
Martinez, A. and A.F. Prosen
1953. Nuevos Culicidos para las entomofaunas de Argentina, Bolivia y Eaaaey:
Mision Estud. Patol. Reg. Argent. 24:27-32.
1955. Las especies de Aedes (Howardina) de Argentina y Bolivia (Diptera, Culici-
dae). Mision Estud. Patol. Reg. Argent. 26:21-41.
1959. Nuevos culicidos para las entomofaunas de Argentina, Bolivia y Paraguay.
Primeras Jornadas Entomoepidemiol. Argent., Buenos Aires, p.603-607.
Martinez, A., A.F. Prosen and R.U. Carcavallo
1964. El alotipo macho de Aedes (Howardina) aurivittatus Cerquiera. Inst. Med.
Reg., An. 6(1/2):53-58.
Martini, E.C.W.
1931. Ueber einige sudamerikanische Culiciden. Rev. Entomol. 1:199-219.
1935. Los mosquitos de Mexico. Mex. Dep. Salubr., Bo!. Tec. (A) 1. 65 p.
Matheson, R.
1950. Medical Entomology. ed.2. Ithaca, N.Y., Comstock. 612 p.
Mattingly, P.F. 3
1955. Mosquitoes (Diptera: Culicidae) from the Tropical Institute at Hamburg.
Roy. Entomol. Soc. Lond., Proc. (B) 24:27-33.
Montchadsky, A.S. and I. Garcia Avila
1966. Las larvas de los mosquitos (Diptera, Culicidae) de Cuba, su biologia y de-
terminacion. Poeyana (A) 28. 92 p.
Peryassu, A.G.
1908. Os culicidos do Brasil. Rio de Janeiro, Inst. Manguinhos. 407 p.
Peyton, E.L., P. Galindo and F.L. Blanton
1958. Pictorial keys to genera of Panama mosquitoes. Mosquito News 15:85-100.
Picado, C.
1913. Les bromeliacees epiphytes considerees comme milieu biologique. B. Biol.
France Belg. (7) 47:215-360.
Porter, J.R.
1967. A checklist of the mosquitoes of the Greater Antilles and the Bahama and
Virgin Islands. Mosquito News 27:35-41.
Prosen, A.F., R.U. Carcavallo and A. Martinez
1964. Culicidae de Bolivia. Inst. Med. Reg., An. 6(1/2):59-124.
Reyes, P.
1955. Lista de mosquitos de Colombia, S.A. (Diptera, Culicidae). Soc. Biol. Bogo-
ta, An. 7(2):46-94.

Berlin: Subgenus Howardina 127

Root, F.M.
1927. Notes on the mosquito fauna of the Republic of Haiti. Amer. J. Hyg. 7:
463-469.
Senevet, G.
1936. Les moustiques de la Martinique. Inst. Pasteur Algerie, Arch. 14:123-134.
1938. Les moustiques de la Guadeloupe. Inst. Pasteur Algerie, Arch. 16:176-190.
Senevet, G. and L. Quievreux
1941. Les moustiques de la Martinique (2e memoire). Inst. Pasteur Algerie, Arch.
19:248-264.
Shannon, R.C.
1931la. The environment and behavior of some Brazilian mosquitoes. Entomol.
Soc. Wash., Proc. 33:1-27.
1931b. On the classification of Brazilian Culicidae with special reference to those
capable of harboring the yellow fever virus. Entomol. Soc. Wash., Proc. 33:
125-164.
Soper, F.L., H. Penna, E. Cordoso, J. Serafim, M. Frobisher and J. Pinheiroz
1933. Yellow fever without Aedes aegypti. Study of a rural epidemic in the Valle
do Chanaan, Espirito Santo, Brazil, 1932. Amer. J. Hyg. 18:555-587.
Spielman, A. and A.E. Weyer
1965. Description of Aedes (Howardina) albonotatus (Coquillett), a common do-
mestic mosquito from the Bahamas. Mosquito News 25 :339-343.
Stone, A.
1963. A Synoptic Catalog of the Mosquitoes of the World (Diptera, Culicidae).
Supp. II. Entomol. Soc. Wash., Proc. 65:117-140.
1967. A Synoptic Catalog of the Mosquitoes of the World (Diptera, Culicidae).
Supp. III. Entomol. Soc. Wash., Proc. 69:197-224.
Stone, A. and K.L. Knight
1956. Type specimens of mosquitoes in the United States National Museum: II,
The genus Aedes (Diptera, Culicidae). Wash. Acad. Sci., J. 14:213-228.
Stone, A., K.L. Knight and M. Starcke
1959. A Synoptic Catalog of the Mosquitoes of the World (Diptera, Culicidae).
Wash., Entomol. Soc. Amer. (Thomas Say Found. Publication 6). 358 p.
Suarez, O.M. and P. Cova Garcia
1961. Estudio morphologio de Aedes (H.) ioliota Dyar and Knab, 1913. Rev. Bra-
sil. Entomol. 10:17-24.
Surcouf, J. and R. Gonzalez-Rincones
1911. Essai sur les Dipteres vulnerants du Venezuela. Primiere Partie. Dipteres
Nematoceres vulnerants. Paris, Maloine. 320 p.
1912. Dipteres piqueurs et suceurs de sang actuellement connus, de la Republique
de Venezuela. Arch. Parasitol. Paris 15:248-314.
Theobald, F.V.
1901. A monograph of the Culicidae or mosquitoes. v.1. London, Brit. Mus.
(Natur. Hist.). 424 p.
1903a. A monograph of the Culicidae or mosquitoes. v.3. London, Brit. Mus.
(Natur. Hist.). 359 p.
1903b. Notes on the genus “‘Stegomyia”’ (Theobald), and its distribution. J. Trop.
Med. 6:237-239.
1905. Diptera, fam. Culicidae. In P. Wytsman, Genera Insectorum, Brussells, Des-
met- Vertemeuil. Fasc. 26. 50 p.

(128 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

1907. A monograph of the Culicidae or a v.4._ London, Brit. Mus.
(Natur. Hist.). 639 p.

1910. A monograph of the Culicidae or mosquitoes. v.5. London, Brit. Mus.
(Natur. Hist.). 646 p.

Theobald, F.V. and M. Grabham
1905. The mosquitoes or Culicidae of Jamaica. Kingston, Inst. Jamaica. 40 p.
Thompson, G.A. |

1947. A list of the mosquitoes of Jamaica, British West Indies. Mosquito News 7:
78-80.

1956. Aedes stenei, a new species of mosquito from Jamaica, B.W.I., with rede-
scription of a closely related species (Diptera, Culicidae). Mosquito News 16:
196-200.

Trapido, H. and P. Galindo |

1957. Mosquitoes associated with sylvan yellow fever near Almirante, Panama.

Amer. J. Trop. Med. Hyg. 6:114-144.
Trapido, H., P. Galindo and S.J. Carpenter

1955. A survey of forest mosquitoes in relation to sylvan yellow fever in Panama

Isthmian area. Amer. J. Trop. Med. Hyg. 4:525 “942.
Urich, F.W.

1913. Mosquitoes of Trinidad. Agr. Soc. Trinidad Tobago, Proc. 13:525-530

(Pap. 574).
Vargas, L.

1956. Especies y distribucion de mosquitos mexicanos no Anofelinos (Insecta,

Diptera). Inst. Salub. Enferm. Trop., Rev. 16:19-36.
Whitman, L.

1951. The Arthropod vectors of yellow fever. In G.K. Strode (ed.). Yellow Fever.

New York, Mc Graw Hill. 710 p.
Whitman, L. and P.C.A. Antunes

1937. Studies on the capacity of various Brazilian mosquitoes, representing the
genera Psorophora, Aedes, Mansonia and Culex to transmit yellow fever. Amer.
J. Trop. Med. 17:803-823.

nARWN

Berlin: Subgenus Howardina 129

FIGURES

Distribution of the groups of Howardina

Distribution of the species of the Fulvithorax Section

Distribution of the species of the Walkeri Section

Distribution of Aedes (H.) sexlineatus

Distribution of the species of the Quadrivittatus, Whitmorei and Aurivittatus
Groups

Distribution of the species of the Allotecnon and Eleanorae Groups

Biting-landing cycle of Howardina species near Porus, Jamaica

Mesonotal scale pattern of Howardina species

Mesonotal scale pattern of Howardina species

Aedes (H.) fulvithorax; pleuron, male genitalia and pupa

Aedes (H.) fulvithorax ; larva

. Aedes (H.) septemstriatus; male genitalia and pupa
. Aedes (H.) septemstriatus; larva

. Aedes (H.) arborealis; male genitalia and pupa

. Aedes (H.) arborealis; \arva

. Aedes (H.) walkeri; adult morphology

Aedes (H.) walkeri; pleuron, male genitalia and pupa

. Aedes (H.) walkeri; larva

. Aedes (H.) aurites; male genitalia and pupa

. Aedes (H.) aurites; larva

. Aedes (H.) grabhami; male genitalia and pupa

. Aedes (H.) grabhami; larva

. Aedes (H.) inaequalis; male genitalia and pupa

. Aedes (H.) inaequalis; larva

. Aedes (H.) stenei; male genitalia and pupa >

. Aedes (H.) stenei; larva

. Aedes (H.) ioliota; male genitalia and pupa

. Aedes (H.) ioliota; larva

. Aedes (H.) cozumelensis; male genitalia and pupa
. Aedes (H.) cozumelensis; larva

. Aedes (H.) albonotatus; male genitalia and pupa
. Aedes (H.) bahamensis; male genitalia and pupa
. Aedes (H.) bahamensis; larva

. Aedes (H.) busckii; male genitalia and pupa

. Aedes (H.) busckii; larva

. Aedes (H.) quadrivittatus; pleuron, male genitalia and pupa
. Aedes (H.) quadrivittatus; larva

. Aedes (H.) lorraineae; \arva

. Aedes (H.) brevis; male genitalia and pupa

. Aedes (H.) brevis; larva

. Aedes (H.) spinosus; male genitalia and pupa

. Aedes (H.) spinosus; larva

. Aedes (H.) sexlineatus; male genitalia and pupa
. Aedes (H.) sexlineatus; larva

. Aedes (H.) whitmorei; male genitalia and pupa
. Aedes (H.) whitmorei; larva

. Aedes (H.) allotecnon; male genitalia and pupa

130 Contrib. Amer. Ent. Inst., vol. 4, no. 2, 1969

48. Aedes (H.) allotecnon; \arva

49. Aedes (H.) guerrero; larva

50. Aedes (H.) eleanorae; male genitalia and pupa

51. Aedes (H.) eleanorae; larva

52. Aedes (H.) marinkellei; pupa —

53. Aedes (H.) marinkellei; larva

54. Aedes (H.) leei; male genitalia and pupa

55. Aedes (H.) leei; larva

56. Aedes (H.) osornoi; male genitalia and pupa

57. Aedes (H.) osornoi; larva

58. Aedes (H.) aureolineatus, pseudodominicii, ecuadoriensis and brevivittatus; male
genitalia

NOILOSS SNLVANITXSS

SNLDJJIAUND

NOILDJSS XVYOHLIA Ns

La 4OWJIUM P

pra y “pe o
: ATS,
9D410UDa|a | ga XD4
a | &< ly ¢
ty —, ‘ :

;———*__ snp aulxas on os
ie UOUDAIOI|D ‘. $0 SK

o,

UHAIN,
%

ESN. - : y >
Se eat. ee Ps ? 4 ak aS
5 ee aN 7
SPSS Ssr— $ npOHALp onb RSE Saree NSS.
nN Ne, he fi
NE eS. nyosng——ti aeuer

NS ES FE TLN,
[Ogg coe SAN

of SSreAY
a : [1a4]DM

3 snjo }ouoq|D —— —
; ; D401)! : :
, * NOILOSS IMSN WWM ets =.
, ay Lu 9 3
bas he “es

SS } i Pop ae
“~ vs
4 ? Ys ree
T ‘ “do ~
i gy ees “ es
Z : 4 £ : Sy
= 2 N
| : Ri ,, ’
. ts “ig =

VNIQUYVMOH

SATIN ALALVLS
00s

0001
SYALIWOTIA

VOIUANY WIGGIN =

e Ps

SXDIOUPAIN #
te)

ie

%

! 2)
i bd
A PP."
r Win
3 =
Sy A k .
TVG OY
t tf
3 Dac)
Sete <=>

snjpeulloounD “Se,
a

Sa TI ——

ool os 0

00s
"GWALAWOTIN

vortanW STC

110})0M a

q2

iN

snip4ouogio :
SNLVLONOS IV

ten ae ,

D{O}]

O!

SATIW ALNLVLS
00s

0001
SYaLAIWOTIA

VOINHNV WIdd

(em eau os

SNLVSNITIXSS —

i

a

SATIW ALALVLIS
90s

Gol 0s 6

ocol 00s

SHALANOTTA

VOIUHINV HIGCIN

os co ‘4 \ | ) 1 c s § oS
ey x VAN 4 -w a - mx

SNJO}JIALApONb
9D9UID140}

lOJOWLIUM

ah
ISYOWLIHM

5)

ne —

IZoullIOW PFs ae
4

sn IND Petes. 7

ID} HAUND™ Petes. FP
JUBPWAUDA “sx

ty

enV LIISA: |

7 :
i ey =

2-S961-1N3-A90100Z-¥19N

SHTIW ALALYLS
00S

SYA LAWOTIA

WV HI

OO os 0
0010s 0
JX
G 2
j
CdIW ae
hes

‘s

SNLDIJIAIAGIG
SISUsIOpDNDA

-199| 2 tox

19|J94uUUDW

10UJOSO
apsoudala

U0UD94Oj|D
DJOWe}onb

NONOSLONNV

| o1osionb

2S SSUNIGHYMOHS)

Hy Qs.
jy ©
Y ee
.: a,
ae CLA!

\

\

\
\
\
\

Toua}s ("H) ‘Vv pue

STjenboeul (“H) Yo We ay
PIedeis (iH) y fo ee tee g

SiInoy UI JWT],

, ie dil ie
\- 4
‘B 7
a ao f
Le Pd fy
\ , / 4 \
\ at / / \
a \ Vv / Py
= PN -.co
\ \ ra ®, 7 <
\ eg ‘a /
\ i” /
\ ]
\ /
{ /
' /
\ !
| /
\ /
\ !
; 1
\ /
1 7®
| a
1 @
1 -
\ /
\ /
\ /
\ ee A
oe:
\/
r)
Z 614

°
e*
e
eo?
°°
e
°°

®
VOIVNVE “SNuYOd UVAN SHIONdS VNIGUVMOH AO ATIOAO ONIGNV'I-DNILIG

0090

00SO

OZ

O€

OS
09
OZ

08

OTT

Ocl

yySneo seojyrnbsow jo JaquinNn

HOWARDINA

acrostichal

Fig.8

inner dorsocentral

outer dorsocentral

lateral marginal

lateral prescutellar

Selle ek ie ne ate oe

ioliota, cozumelensis
albonotatus,bahamensis,

lateral marginal

median prescutellar

antealar

supraalar

busckii quadrivittatus lorraineae

HOWARDINA

‘@ ® °° =
5 sa Q
»

& 9° ne
a a 2)
= o ”

@ ro)
= >
2
B
oO
z
=
8 =
= os
~~ @D
oo
;
w ~ £ —_
o ®
> ~~ =
oO mo @
=<
@ _> =<
a Co =
poe 2 yas , a
x< co ro]
> OE S
—
= Qo =}
3 Bo
5 = s
. f: oD
g ‘bb
5 Fl
Ss 8
3 a
8 8
3
Ww
” = =
—
= & <<
2 re =
2 o 2
fo% an =
wo oO ®
) —
g : :
@ @ =.
a Qa. r)

inezi

mart

ittatus

agurivt

iensis

ecuador

HOWARDINA

=e
if \\( Wy, sey

[ |
oN

fulvithorax

_TR 591-121
Monos Island
Trinidad

HOWARDINA

6 apes mm if

mi

fulvithorax

5 4 TR 680-102
MG \K Verdant Vale
Arima, St. George
Trinidad

~HOWARDINA -

cy
septemstriatus

HOWARDINA

septemstriatus
PA 301-107
Almirante
Bocas del Toro
Panama

HOWARDINA

ee,

arborealis

=e.

>)

HOWARDINA

Aloesoebanja
Upper Surinam
Surinam

Bis, pa arborealis
11

HOWARDINA

— —— pops CLeoiitkhiadaL
COSSSSSS SSS 2 2222neee2elZ0eZeee 24
SAAR PPLE LE SELES SESS SESESE

SSS \y
— Ry
= “as

postgenial plate

4 i 2
ee
1
fa ,

< pate

IX tergite

spermatheca

VIII sternite

D>

Co ue
See

walkeri ~ 4'o

EO iE, a
o 2 |
IGG 6

HOWARDINA

walker}

JA 379-10
Mt. Diablo

amaica

HOWARDINA

i 10 7 4 3 . "Gi :
4

)
Vill

booon abMRR: ‘At
i iy =
; : L
A),
ie » 2
R/
y Wy Wy yy |
: ths A ‘ Nii
Man) Mn \ ai
Lp |
y hy

iV yyy
] Ny hy
Lani

walkeri

JA 243-25
sd, i) Porus
Z| Hf \\N Manchester
I 4 AN Jamaica
5 {|

HOWARDINA

| vie Fig.12

« x
\ WwW
S WW 1.0
r JA 682-101
aur t es Newcastle

St. Andrew
Jamaica

HOWARDINA .

3 Wy *

0
4
sp
9
8
2
13 BRM gua css rN
Bs 50 1
Proximal Distal Att 7
PT Ae ‘ M :
‘ 14 «
VIII
: \

JA 682
Newcastle
St. Andrew
Jamaica

aurites

HOWARDINA

JA 632-103
Porus
Manchester
Jamaica

HOWARDINA

grabhami

eu
TUS
WS WY 2
pass "

Nae

JA 290-11
Stony Hill
St. Andrew
: , St An
inaequalis amaica

HOWARDINA

Temple Hall
St. Andrew
Jamaica

inaequalis

HOWARDINA

stenei Hope Farm

Porus, Manchester

HOWARDINA

=

Iii
x 4
4 :
ee \Q H\\ Ih
; tf / \ Hh. \
NaN
Z \
I

\\

\

a xf

stene|

\V4jI \ A

, y
\y & ¢ i,
ae Kanne

St. Andrew

\
TR 940-111 *
Trinidad

Arima

HOWARDINA

ioliota

HOWARDINA

TR 938-129
Arima

St. Andrew
Trinidad

HOWARDINA

Island of Cozumel

Yucatan
Mexico

cozumelensis

HOWARDINA

cozumelensis

HOWARDINA \\

Fig. S| re

San Francisco Mts
Dominican Republic

albonotatus

HOWARDINA

bahamensis

HOWARDINA

BAH 1

Hunter

Grand Bahama
Bahama Islands

bahamensis

HOWARDINA
Fig. 34

Aedeagus

Guadeloupe population

DOM 41-10

“s ; sak
busckil St Pat

Dominica

HOWARDINA

busckii

0 HOWARDINA
fate Fig. 36

HOWARDINA

mee 3)

Fy A. |

fl, <j
CS PT

VU

S$ ck
~ thhaa 9 to edad
i nM (eS
Al W
a Za
4 r Y X
fg \

—s ~
— = =
—————

\

quadrivittatus

CR SI
La Sierra
Costa Rica

HOWARDINA

MEX 362
Omilteme
Chilpancingo
Guerrero

lorraineae

Mexico

HOWARDINA

HOWARDINA

Ec

UNS
i it Mh \

}
HY

|
9 ij it i
ny ;

fh

MP

Cs 1M

Chiriquisoto
Bocas del Toro
Panama

brevis

HOWARDINA

Fig. 4
\

us

\ ,
i. \\ ‘
Ne ‘
* |? J
ie rol
1
CR 13-111

Cervantes

spinosus Cartago

Costa Rica

HOWARDINA

5 VItl

CRAB=LIT

Cervantes .
Cartago SpInosus

Costa Rica

HOWARDINA

aN

\,
c7 i ie

TR 1041
La Laja Trace
Arima

sexlineatus pont

HOWARDINA

\
i Mt

WH

~X
Pe
= 2 2 =
=
“te
SS

Dat
NY
Nee
Rees
<

TR 1099
Verdant Vale
Arima

St. George
Trinidad

sexlineatus

HOWARDINA

=

——

!
/
/

Zz
a SaaS ee

ANN # = =
SSS Onn : : s
SS NS | SS :

og

COB 67-25
illavicenct

whitmorei

HOWARDINA

COB 68 ; 5
Buena Viste whitmorel
Villavicencio, Meta

Colombia

HOWARDINA

CR 13-102
Cervantes

Cartago

allotecnon Cosa Riow

HOWARDINA

CR 14
Cervantes
Cartago
Costa Rica

allotecnon

HOWARDINA

MEX 429

Vivero Rancho Teja

Ixtlan de Juarej Q uerre ro
Oaxaca
Mexico

HOWARDINA

.
Ny, , "
MINN eae
COB 10-22 ;
Buritica

ntioquia
eleaqnorae Colombia

HOWARDINA

COB 10-22
Buritica,
Antioquia
Colombia

Ans iK~ eleanorae

HOWARDINA

COM 63-15
Suba
Bogota, Cundinamarca q

marinkellei Begoten.f

HOWARDINA

<< aS ee ais ed

COM 63
Suba
Bogota, Cundinamarca
Colombia

Sa

marinkellei

HOWARDINA

COL 16-100
Purace
Cauca
Colombia

HOWARDINA

COL 54
Paramo de Purace }
Huila e e
Colombia

HOWARDINA

Fig. 96

COB 34-15

Chorro de Padilla
Bogota, Cundinamarca
Colombia

osornoi

HOWARDINA

wy l

II

~o

osornol

HOWARDINA

aureolineatus pseudodominicii

Yr
fogs
i

Wy Lin

SS

7 | Fig.58

brevivittatus _ ecuadoriensis

Berlin: Subgenus Howardina 189

INDEX TO SCIENTIFIC NAMES

Names of valid taxa included in Howardina are in boldface type, names of valid extralimital
taxa are in roman type. Italic type is used for names of all taxa considered invalid in this work
(synonyms, homonyms, misidentifications, gross erroneous spellings, and so on). Misidentifications
or invalid taxonomic interpretations are indicated by the use of the term “of authors” or “of”
followed by a specific author, and are listed as separate entries following the entry for the cor-

rectly applied identical name.

Boldface numerals refer to the principal text reference, italic numerals to figure numbers, and
roman numerals to secondary text references, with the suffix k indicating mention in a key.

Aedes Meigen, 1, 8, 9

Aedini, 8

aegypti (Linnaeus), 63

albonotatus (Coquillett), 29k, 30k, 31k, 54,
58, 59-61, 62, 63; 3, 8, 31

albonotatus of authors, 61

Albonotatus Group, 7, 27, 29k, 30k, 31k, 53,
58-59; 1, 3

allotecnon Kumm, Komp and Ruiz, 10, 74k,
75k, 76k, 98, 99-102, 103, 104; 6, 9, 47,
48

Allotecnon Group, 7, 70, 71, 73k, 76k, 87,
93, 98, 101, 104, 105; 7, 6

arborealis Bonne-Wepster and Bonne, 5, 9, 14,
15k. 16,.19, 20, 21, 22-25, 26; 2, 6: 14,
15

arborealis of authors, 19

argyrites Dyar and Tunez Tovar, 28k, 29, 30,
31, 32, 33, 35, 37-38; 3, 8

aureolineatus Berlin, 14, 15k, 21, 24, 25-26;
2, 8, 58

aureostriatus (Grabham), 42, 43, 44, 48

aureostriatus of authors, 46

aurites (Theobald), 28k, 29k, 30k, 31k, 39,
40-42, 43, 44, 46, 47, 48, 49; 3, 8, 19, 20

aurites of authors, 43, 46

Aurites Group, 6, 7, 8, 26, 27, 28k, 30k, 31k,
33, 35, 38-39, 42, 44, 45, 67; 1, 3

aurivittatus Cerqueira, 73k, 74k, 75, 118-119,
120;5, 9

aurivittatus of authors, 119

Aurivittatus Group, 7, 70, 71, 72k, 105, 117-
118,120; 1, 5

bahamensis Berlin, 6, 12, 29k, 30k, 31k, 59,
60, 61-63; 3, 8, 32, 33

brevis Berlin, 72k, 74k, 75k, 76k, 77, 78, 79,
81, 82-84, 85, 86;5, 9, 39, 40

brevivittatus Berlin, 73k, 74k, 75, 105, 116-
117; 6, 9, 58

busckii (Coquillett), 3, 9, 12, 28k, 29k, 30k,
31k, 64, 65-70; 3, 8, 34, 35

busckii of Hill and Hill, 67
Busckii Group, 7, 8, 26, 27, 28k, 31k, 32k,
58, 63-64; J, 3

caledonicus (Edwards), 67

cozumelensis Diaz Najera, 29k, 30k, 31k, 53,
54,55, 56-57; 3, 8, 29, 30

Culex Linnaeus, 22

delpontei Martinez and Prosen, 96, 97

dissimilis subgroup, 8

dominicii Rangel and Romero-Sierra, 3, 8, 11,
87, 88, 90

dominicii of authors, 93, 95, 96, 97

ecuadoriensis Berlin, 70, 73k, 74k, 75, 90,
104, 105, 115-116, 117; 6, 9, 58

eleanorae Berlin, 73k, 74k, 75k, 76k, 104,
105-107, 108, 110, 111, 112, 113, 115,
116; 6, 9, 50, 51

-Eleanorae Group, 7, 71, 73k, 76k, 80, 104-

105, 109, 118; 1, 6

Finlaya Theobald, 3, 7, 8, 14

fulvithorax (Lutz), 5, 8, 9, 14k, 15k, 15-19,
20, 21, 22,23, 2425:2, 810-11

Fulvithorax Group, 6, 7, 26, 64

Fulvithorax Section, 4, 5, 6, 7,8, 12k, 13-14,
64; 1,2

grabhami Berlin, 26, 28k, 29k, 30k, 31k, 39,
40, 41, 42-46, 47, 48, 49, 51; 3, 7, 8, 21,
2A

greenii (Theobald), 3

guatemala Berlin, 73k, 74, 75, 98, 103-104;
6,9

guerrero Berlin, 73k, 76k, 98, 100, 101, 102-
103; 6, 9, 49

Haemagogus Williston, 3, 8, 22, 27
Howardina Theobald, 3, 4, 6, 7, 8-11, 13, 14,
27, 28, 35, 87, 93, 104

190 Contrib. Amer. Ent. Inst.,

inaequalis (Grabham), 28k, 29k, 30k, 31k, 39,
40, 41, 42, 43, 44, 45, 46-49, 51; 3, 7, 8,
23, 24

inaequalis of authors, 40, 43

ioliota Dyar and Knab, 29k, 30k, 31k, 53, 54-
56,57; 3, 8, 27, 28

Ioliota Group, 7, 13, 27, 28k, 30k, 31k, 33,
52-54, 58, 64; 1, 3

Iquitos form (sp.5), 14, 15k, 25, 26; 2

leei Berlin, 73k, 74k, 75k, 76k, 106, 108,
109, 110-112, 113, 114; 6, 9, 54, 55

leucocelaenus Dyar and Shannon, 8

lorraineae Berlin, 72k, 76k, 77, 78, 80-82, 85;
5, 6.00

marinkellei Berlin, 73k, 75k, 76k, 107-110,
111, 112, 113, 114; 6, 9, 52, 53

martinezi Berlin, 73k, 74, 75, 118, 119-120;
5,9

osomoi Berlin, 73k, 75k, 76k, 107, 108, 109,
110, 111, 112-115; 6, 9, 56, 57

palliatus (Coquillett). 15, 16, 18
_ pseudodominicii Komp, 6, 72k, 74k, 75, 93,
95, 96-98; 5, 9, 58

quadrivittatus (Coquillett), 8, 72k, 74k, 75k,
76k, 77-80, 81, 82, 83, 84, 85, 86; 5, 8,
36, 37
quadrivittatus of authors, 105, 108, 110, 112,
114,116
Quadrivittatus Group, 6, 7, 71, 72k, 75k, 76-
- 77, 80, 87, 98, 101, 105; 7, 5

Sabethes Robineau-Desvoidy, 22

vol. 4, no. 2, 1969

septemstriatus Dyar and Knab, 14, 15k, 16,
17, 18, 19-22; 2, 8, 12, 13

septemstriatus of authors, 23, 24

sexlineatus (Theobald), 3, 11, 72k, 74k, 75k,
76k, 87-92, 95,97;4, 9, 43, 44

sexlineatus of authors, 118

Sexlineatus Group, 6, 7, 70, 71, 72k, 76k, 86-
87,93; 1, 4

Sexlineatus Section, 4, 5, 6, 7, 11, 12k, 13k,
27, 28, 33, 39, 70-71, 90, 93; 1

Soperia Komp, 3, 4, 8, 11, 93

species 5, 14, 15k, 25, 26;2

spinosus Berlin, 72k, 74k, 75k, 76k, 77, 78,
79, 81, 84-86;5, 9, 41, 42

Stegomyia Theobald, 3

stenei Thompson, 28k, 29k, 30k, 31k, 39, 40,
41, 42, 43, 44, 45, 47, 48, 49-52: 3; 7, 8,
25, 26

tachirensis Anduze, 15,18 —

terrens (Walker), 22

thaxteri Dyar and Knab, 65, 67

tracei Senevet and Quievreux, 3, 65, 67

vanemdeni Martini, 72k, 74, 75, 118, 120; 5,
9

walkeri (Theobald), 3, 6, 8, 11, 27, 28k, 29k,
30k, 31k, 32, 33-37, 38, 93; 3, 7, 8, 16-18

Walkeri Group, ‘7, 8, 26, 27, 28k, 30k, 31k,
32-33, 38, 39, 53, 64; 1, 3

Walkeri Section, 4, 5, 6, 7, 8, 11k, 12k, 13k,
26-28, 39, 93; 1

whitmorei Dunn, 3, 6, 10, 72k, 74k, 75k, 76k,
92, 93-96, 97;5, 9, 45, 46

Whitmorei Group, 6, 7, 10, 11, 12, 70, 71,
72k, 76k, 87, 92-93, 105; 1, 5

isai'h

et EAT

(Continued from inside front cover)

Parts of Volume 4, with prices

No. 1. Mosquito studies (Diptera, Culicidae), 1968.
X. Peters, T. Michael. Dixinae originally described from North America. 7 pages.
XI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann. Mosquitoes
originally described from Argentina, Bolivia, Chile, Paraguay, Peru, and
Uruguay. 29 pages. Price: $1.25, postpaid.

No. 2. Mosquito studies (Diptera, Culicidae), 1969.
XII. Berlin, O. G. W. A revision of the Neotropical subgenus Howardina of Aedes.
190 pages, 58 figures. Price: $6.00, postpaid.

Contributions
of the

American Entomological Institute

Volume 4, Number 3, 1969

my"

BIOLOGY, TAXONOMY, AND EVOLUTION
OF FOUR SYMPATRIC AGRILUS BEETLES

(COLEOPTERA: BUPRESTIDAE) “~.
MOD aes
‘Robert W. Carlson

and Ae |
Fred B. Knight

CONTRIBUTIONS
| of the
AMERICAN ENTOMOLOGICAL INSTITUTE

The Contributions are for larger papers on insects. Each paper is a separate number, with
separate pagination and index. Separate numbers aggregating about 500 pages constitute a volume.
Issues appear irregularly, as suitable manuscripts are available.

Copies are sold separately or in subscriptions to complete volumes. Complete volumes are
$12.00. The price of separate numbers varies. Subscribers are billed for each volume with its
beginning number, and receive the parts as issued. Orders for separate numbers that total less
than $8.00 must be accompanied by payment. Address orders or correspondence to the American
Entomological Institute, 5950 Warren Road, Ann Arbor, Michigan, U. S.A. 48105

Parts of Volume 1, with prices

No. 1. Dasch, Clement E., 1964. The Neotropic Diplazontinae (Hymenoptera, Ichneumonidae).
75 pages, 69 figures. Price: $2.25, postpaid.

No. 2. Mosquito studies (Diptera, Culicidae), 1965.
I. Belkin, John N., etal. A project for a systematic study of the mosquitoes of
Middle America.
Il. Belkin, John N., etal. Methods for the collection, rearing and preservation of
mosquitoes.
78 pages, 4figures. Price: $2.25, postpaid.

No. 2a. Same as no. 2, but in Spanish. Price: $2.25, postpaid.

No. 3. Matthews, Robert W., 1965. The biology of Heriades carinata Cresson. (Hymenoptera,
Megachilidae) 33 pages, 23 figures. Price: $1.00, postpaid.

No. 4. Mosquito studies (Diptera, Culicidae), 1965.
Ill, Ramalingam, Shivaji, and John N. Belkin. Two new Aedes from Tonga and Samoa.
IV. Belkin, John N. The mosquitoes of the Robinson-Peabody Museum of Salem
expedition to the southwest Pacific, 1956.
34 pages, 3 figures. Price: $1.00, postpaid.

No. 5. Mosquito studies (Diptera, Culicidae), 1965.
V. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.

Mosquitoes originally described from Middle America. 95 pages. Price: $2.75 postpaid.

No. 6. Mosquito studies (Diptera, Culicidae), 1966.
VI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.

Mosquitoes originally described from North America. 39 pages. Price: $1.00, postpaid.

No. 7. DeLong, Dwight M. and Paul H. Freytag, 1967. Studies of the world Gyponinae
(Homopotera, Cicadellidae). A synopsis of the genus Ponana.
86 pages, 257 figures. Price: $2.50, postpaid.

No. 8. Contributions to the mosquito fauna of southeast Asia, 1967.
I. Delfinado, Mercedes D. The genus Aedes, subgenus Neomacleaya a Theobald in
Thailand. 55 pages, 20 figures. Price: $1.75, postpaid.

Parts of Volume 2, with prices

No. 1. Contributions to the mosquito fauna of southeast Asia, 1967.
II. Bram, Ralph A. The genus Culex in Thailand (Diptera: Culicidae).
296 pages, 104 figures. Price: $9.00, postpaid.

No. 2. Schuh, Toby, 1967. The shore bugs (Hemiptera: Saldidae) of the Great Lakes region.
35 pages, 30 figures. Price: $1.25, postpaid.

No. 3. Snider, Richard J., 1967. The chaetotaxy of North American Lepidocyrtus s. str.
(Collembola: Entomobryidae). 28 pages, 57 figures. Price: S106. postpaid.

No. 4. Contributions to the mosquito fauna of southeast Asia, 1968.
Ill. Delfinado, Mercedes D. The genus Aedes, subgenus Neomacleaya Theobald of
southeast Asia. 76 pages, 26 figures. Price: $2.25, postpaid.

No. 5. Contributions to the mosquito fauna of southeast Asia, 1968.
IV. Knight, Kenneth L. Species of the subgroup Chrysolineatus of group D, genus
Aedes, subgenus Finlaya Theobald. 45 pages, 12 figures. Price: $1.50, postpaid.

Parts of Volume 3, with prices

No. 1. Mosquito studies (Diptera, Culicidae), 1968.
VII. Belkin, John N. The Culicidae of New Zealand.
182 pages, 30 figures. Price: $5.50, postpaid.
No. 2. Mosquito studies (Diptera, Culicidae), 1968.
VII. Zavortink, Thomas J. A prodrome of the genus Orthopodomyia.
221 pages, 41 figures. Price: $6.75, postpaid.
No. 3. Fleming, Richard C., 1968. Head musculature of sphinx moths (Lepidoptera:
Sphingidae). 32 pages, 37 figures. Price: $1.00, postpaid.
No. 4. Mosquito studies (Diptera, Culicidae), 1968.
IX. Belkin, John N. The type specimens of New World mosquitoes in European
museums. 69 pages. Price: $2.00, postpaid.

(Continued on back cover)

Ob
“gle

Contributions

of the

American Entomological Institute

Volume 4, Number 3, 1969

me

BIOLOGY, TAXONOMY, AND EVOLUTION
OF FOUR SYMPATRIC AGRILUS BEETLES

(COLEOPTERA: BUPRESTIDAE)
by
Robert W. Carlson
American Entomological Institute, Ann Arbor, Mich.

and
Fred B. Knight

Department of Forestry, School of Natural Resources
University of Michigan, Ann Arbor, Mich.

ii

TABLE OF CONTENTS

DE a se ee le ae as oo Oe aia ay yet uy ah cal ig Vv
Piet A Rg Ce a ea ee ee a ey eg ae eee a eS vi
Dr EC Ee ee ie oe i Gea EN Bower Re Oy a a as 1
ee er ee a ee ee 1
PEG OURO UOU ee Ge a ee a a ee a 1
PU RN BO Ce PORER ee eee eee ee eS 1
THE BON POPPA ORR aa ee Pe ee OS 2
AEE OOP CIRC eee ee ae 2
Se Uh I Pee a a ae
OE a eee oka ie le eG ee are Cele k at gras 3
The Gee Oot a es Ga ae ee 3
ey A RP RTA i hi ee ae es eae a 3
ee a ee Sa eee 4
Ee Arie GORIMe COM Ck ke aN eek ee ee oe eG 4
Oe ee a eae So 4
PE a ete hye ee ee 4
WORT HOLOGIC ANAT VSG ee oe ee 5
a ee ee ee ca
ee ee PN a 12
Fe Ge cae Ae ee WMS a tents 13
a cu ay acme. co a a Pe ae ae gerard am 14
GENERAL BIOLOGIES OF AGRILUS ANXIUS, A. PENSUS,
A, HORNL AND A. GRANULATUS LIRAGUS... 5. eee es 14
a A ce a a ae es de ae 14
Oe OO BC BORER ee ee ee ei ee 14
Te ONE POCA BORER yee ee Se eee ec ee 15
erE AE ROE GOR ee a ile uke) woe te Ses ue a an 15
PO ae a a 15
EEE Oe Oe a RO ee EE ROME MRR gr gle eR IR eile acs 16
AN a a eS a 16
a ree es ee a ea 16
ee Ae ee NG VAN Satta ug Wy Cais ung 17

CONTENTS | ili

HATCHING OF BAAS e205 eer We ea Mele eer wyl nae Bt, vee terawaug ee oy 6 ae
LARVAL PEE DING PERIQD 3 tii 6 ferent ee A a le eae OG 17
The Bronze Birch and Bronze Poplar Borers 4 win 4% ence a we 60 18
The Alder-Birch Bover 3.000. Vai ha faye Gon Bisa we soe i. 18
The Aspen Root Girdler so 0.4) 5 ee aaah ee i ee 18
POST-FEEDING LARVAL PRION 645 1 21
Formation of tie Pupal Chamner oo Gg aad a ie 21
DIADAUSE ee Ae re ete Pig pigsty gE gi GN OU L: iggiiel cabre auaae 22
PU PAL) TIRAINB CN i ee ge ie ati al as Meee ree ee Lee aa
IO LD: BEGIN i ee ee a! ah ic wis wreaks 8 amare es a ae are he
PUY POCONO ii ce ea ae gh ae hie on a ae en a ae eG 22
Resulis mad Disondeiin cs i. 8 ee ae Ge $4.
Population Regulating Factors. 6.9 ee ee ese iw alin me os 26
HOST SOIT se UB es a ewe ea ONY are MI hE 8, 26
The Bronze Birch and Bronze Poplar Borers............. 26
Dae Pide re DITO BOON 6 i kk a a is ee 31
TPne ABP EO HOOU GICCLET oo ai aaa a et a ek 31
DIOR WOMB RC Gd: Fe AR a We ee Ld sp ge ae oe agelige ea a a tie iN 37
ATR a ep eee a |) pee ahies ieee 37
tat BEAVER ee ee Gas 40
Bee BN WEAPON 6) on wn 4 ined nh ke eel wel eat ae tate? 6 40
DigGe veh ee ee hs ea a Gee ae Ae ee ee 41
Host Beleeiie ae ea 41
ARTIS AND G. TARAGUS by does ai ume a ai rime liars te .’, 42
TE ea ke ie ew Say ele Cie te ee Ns ee ae a ade Mana Oe emu RH aes 43
PEC TIOUS BU id ny eee alan eM MO VL a! Buhl VLG aT ae amt 44
DUO Ce is ye ae yk AO MA Le are 45
hes oe: ee ce nen TPR Ue MME E MC ec LUM pe cer ar 45
BFS CU BI ik aie ela ime ete a it Nea ag eat Grae argtae i as hak a 48
POE TAN 50 bce Rah on aah al oe A ARR AG) ig geo MR etal Dela s}9)
boa ae Ne oD ie EC i 8 em ct ne Pe Cac PAU DION ars VASE Ce ga 2) 55
Fd cick 4. EEA CRE Pa ie Gel CR ITORUL OUR Cony Li vair OU Ay UES Cees PEE Moa PD a 25
Pemele Organs of Penrose i se ities all a ae 06
BAY OE, CO EA Te ice oe te ae gL AON I TG aT ie Mie ea ds TTA alae 58

ELEM ELL 4 UE CEL Ua Pen MPa ma NT CRC ROAST TOUU NOISE MUA a BS Seti Rg 58

iv CONTENTS

Insertion of Aedeapis aad Insemination. ee 61
Behavior of Males Dnder Natural Conditions 3) 9 Pah a 61
Bronze Birch and Broaze Poplar Borers... 2°.°... 62

Mie Aeron toot Gir die r .-. 5 ee ST eee a. 65

PRN te AeCEPIANCE GF MALES. 2. we a ae. Bae eee A we 66
Th i oe ONY A a i as ee RE RO PO Ee eh, 69
ee Sk ae ee dm a A a a De 71
ee SI i eke ke ee ee ee ee SER 71
a Oa ee ie ia) ete Se Se aks ee RS eae ee ae iy 71
Fe ee tea ae Pe ae ee ae ag a ee ge Re a op 73
Pee Pe re he ee ee ee eee 74

ee Oe eat Peed ated he ot CR Oe en ee 18

he Ame Betula DOrers oe) a eA ee AOE a 80

Oe Ie er Se a ee a Ee et Od So 81
ee as Pe ee See as cei eee AON ON aoe Siti od 82
Oe ON eee ee ge eh Re es A Ae ee SES ee ea 82
itect Ge Olisalic Chanee x be a kt ek OO ee Ge 82

De i te te icra fete te tele ata Seer eee ae Om 82
re a a a aoe hak ae Te Pe Se eae Te ae ta ioe tea le Le 84
ape) COMnCe One. fo be bs hee i et ee a a Ge a 84
hier at oh gh a hind aoa tow? Oi alsin lS Ret Meee hs Sb ee 84
ieee ation Gl GDECIES a aa ka es ee ie eho te i 87
Srecistiot in the: Busines Oreup cad as « SRE ee 87

Tee Sretine Species Concept. Lee eis ee Or, 89
Be OS i i a Pe ahah ON ah a eatiat oe al ge Wa 92
re a RR ec EM Ny ta Marta tp hp te aegetatne MeL EES aceM y 93
Pe ee Nea ee Rok ee kA a A oe 95
TO tery is Behe tee hee a ee tate Sas a Nal Beles es aes eg whe 96

LIST OF TABLES

Average deviation from long-term

mean air temperature for Stambaugh, Michigan. ...........

Age-class frequency distribution of

healthy and agrilus-killed Populus tremuloides suckers........

Development of Agrilus horni larvae

in selected Populus tremuloides root suckers’ 2°. Po ee ee a

Attractiveness of various types of

Populus tremuloides host material to Agrilus granulatus HiPagus se

The occurrence of Agrilus horni eggs in the Kenton and Phelps
areas on Populus tremuloides root suckers in height classes of

Chi-square comparison of observed and
expected size-class frequency distributions of Populus tremuloides

suckers bearing Agrilus horni eggs in the Kenton and Phelps areas...

Chi-square comparison of number of Agrilus horni eggs above and

below the 1 inch level on the bases of Populus tremuloides suckers. .

Position and situation of Agrilus horni eggs on Populus tremuloides. .

Chi-square comparison of positions of

Agrilus horni eggs on smooth bark with positions of other eggs... .

.93

ays)

vi

oon oaounrtrt why kr

Oa
Swe AaAnNoaurkhwnro

Do Nw bd
Com

iw)
os

20

26

27

28
29
30
ol
32
33
34
30
36
ov
38
39
40)
Al
42
43
44
45
46
47

LIST OF ILLUSTRATIONS

ROOT VG etre A eae! wl elbel eg woe 4 koe ee 5)
Dorsal surface of the pygidium of an Agrilus granulatus liragus?...... 9
Dorsal surface of the pygidium of an Agrilus anxius ¢............ 9
Elytra of Agrilus granulatus liragus, A. anxius, and A. pensus....... 10
Aeteasis OF AGT Ins anus (VenITAl GHC) Co ce ae es 11
Aedeagus of Agrilus granulatus liragus (ventral side). ........... H1
PeGexgus Of Agr ibis ROral (VENUE! SIGE) eee ke ee Ae ae 11
eueees OF AST ile pense EVENIFal GUIe). a oak eee ee eae 11
ee a a ee Ua aie Mahim Tuas CW GAL ied ki a Sih NRE AW ae 12
Labi GO Aerie Stanulatis Lirsgeus IerVva i ee ee eee ae 14
ee ee cr Se a ew | ale pues Ns ee ea ea wes ew) a tae a a 18
PT, PEI POE ee Bln ce eee eek ce ere ‘4s
Small stapled-on emergence cages for.A. ANKIUS 2. ....6 cb beep ecle eee ee « 23
Peer ite Wie tt her BON Ce, BO OE ie i cad Ws ne ak ioe pci cons epee, He HEE ee Mie ie 24
Adult emergence patterns, A. horni, A. anxius and A. g. liragus...... 25
interaction of Agrilus gallery and pRIOeGM, MOMHIRe: 240.006. feces andes pAee * 30
Unassembled cheesecloth cage for Agrilus horni oviposition. ........ 34
Assembled cheesecloth cage for Agrilus horni oviposition... ... 2.2.6. 34
MW OOCeR INOUNTIOTR TOP REPS CORR od i uk ehh ae mg ae Mell eap aca eke iam cinbaiee > 35
Density of horni-infested suckers as function of sucker stand density

Aare POWIet. TUG pr ae a ue Do Moa bs ge eice es ue ale ie eee ae
pitdy ares. of Nord (1962) and Fowler (1) 9G) oo ses bce, Wilke (sos euvin suty depron Com Be 46
Agrilus horni study area near Beechwood, Michigan. ... . «+ ce. + aoe) 46
Ratio of observed to expected numbers of horni egg-bearing suckers per
height class plotted. as.atunction of sucker Weight ows ace scenes eevee! 51
Ratio of observed to expected numbers of horni egg-bearing suckers per
diameter class plotted as a function of sucker diameter. .......... 51
Ratio of total eggs to expected number of horni egg-bearing suckers per
height class plotted as a function of sucker height. ............. 52
Ratio of total eggs to expected number of horni egg-bearing suckers per
diameter class plotted as a function of sucker diameter. .......... 52
Number of suckers and cumulative number bearing Agrilus horni eggs

per hypothetical random Sample. Of LO0OE iss ew acd ecces cs ircemese cae be 04
Aedeagus, Agrilus granulatus liragus (dorsal side). ...........-. 56
Ovipositor of Agrilus anxius in extended and retracted positions...... 57
Probable internal position of A. anxius aedeagus during copulation. .... D8
oii & Ie rtiat OFeans Ol FEDYORUCTION O'S a eee kw ee es 58
ORCI EO ee te aan dae aloes ab a: dnibeaitin gh RU 59
IC et A VR ree oe a tac acins eoclns cdesie anova asa tmanas 59
A)! SRS tte IROL On OL HOCEARUB ee ee eh ae ee eee 61
Se, Cf eT WEMETAIE ge kt ee Ma eee ek a ee, bei we cae 63
OUI GOVT IICre LORCING Blam ela pa, Wo ay oe ne we ee ee lade 63
Miele Geee TOT ODBOTVIN Aer US OTR Cae ake ee lee ee wh eG 66
Antennal fluttering by an Agrilus anxius CS incopulo............. 67
Antennal stroking movements by an Agrilus granulatus liragus “ in copulo . 68
Striate areas of scape and front (Agrilus vittaticollis %)........... 69
SSE EO OE eee CLUE PEED a eK aeRO mes ao MCR A EO Re oe 70
Internal surface prosternal lobe in Agrilus horni. ........4.4+.-+.2--. 71
Audiospectrograph of stridulation by Agrilus granulatus liragus %..... 72
Phyletic tree for part of the Agrilus anxius group. ........2454-+ 26s 75
Penis Sper OF ee OPPO kV elie el pa SR ORC a ate ee 79
Elytra of Agrilus ater, A. guerini, and A. quadriguttatus niveiventris ... 79

Mayr’s model for evolution of host specific sibling species. ........ 91

INTRODUCTION
Objectives

Beginning in the summer of 1964 we endeavored to study the natural history
and taxonomy of four sympatric Agrilus Megerle’ (Coleoptera: Buprestidae).
These Agrilus were of particular interest because of their morphological sim-
ilarities. Three of them, the bronze birch borer (Agrilus anxius Gory), the
bronze poplar? borer (A. granulatus liragus Barter and Brown) and the aspen
root girdler (A. horni Kerremans) had only been distinguished rather recently.
The fourth species, ‘ee pensus Horn (= betulae Fisher) is distinct from the oth-
ers in coloration, but very similar otherwise. The credibility of contentions
that these taxa are very closely allied will be deliberated here in the light of:

1) A taxonomic study of these and related Agrilus, which included working
both with specimens and literature.

2) An analysis of their larval habits, their ecological relationships to their
hosts, and their population-limiting factors. Much of what is known about
these aspects is taken from the literature and reinterpreted in the light of more
recent findings, and pertinent writings on other Agrilus are also discussed.

3) Direct and indirect observations on the behavior of the adults in the
field and laboratory.

4) A preliminary study of interspecific mating and a theoretical analysis of
phylogeny, zoogeography, and speciation (or subspeciation).

Historic Background

THE BRONZE BIRCH BORER. A. anxius is probably the most notorious Ne-
arctic species in the genus, and at one time was thought to be the cause of birch
dieback. It was first recognized as a pest of shade trees and ornamental birches
in the late 1800’s and was called ‘‘the bronze birch borer’’ by Chittendon (1898).
Chittendon says that ‘‘one correspondent has expressed the belief that if radical
measures are not adopted, the loss of every birch in the city of Buffalo in the
near future is imminent. This insect has already destroyed the common white
birch [Betula papyrifera| and... many of the cut-leaf and European white bir-
ches. It even attacks trees planted but a year before. ’’

Most of the reports of damage came from the northeastern U. S., perhaps
because more birches had been planted there as shade trees. Nevertheless,

‘The majority of authors credit Curtis (1825) with originating the name Agrilus. How-
ever, Megerle (1823) published a price list in which the heading ‘‘Agrilus v. M. Bupres-
tis)’’ appears. This simple listing credits Megerle with authorship, and Curtis (1825)
merely designated one of Megerle’s included species, Buprestis viridis L., as the geno-
type.

a
The term poplar will here be used to refer to species in all sections of the genus Populus.

3 ;
Described as Agrilus liragus Barter and Brown (1949); status changed by Carlson (1969).

2 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

the problem was not confined to that region. Chittendon (1900), for example,
notes that in 1899 ‘‘Prof. F. C. Newcombe of the University of Michigan, Ann
Arbor, Mich., wrote that what was without doubt this species, and which he
designated as the white birch borer, had been in that locality for two or three
years and had killed half the white birches in the city of Ann Arbor. ’”’

Swaine(1918) first reported anxius killing trees in forests. This was the
start of a devastating and puzzling epidemic of birch dieback which primarily
affected Betula papyrifera, although some mortality of Betula lutea (= alleghani-
ensis) was also ascribed to this affliction. By 1951, 67 percent of the birches
in Maine had been killed, and 15 percent of those remaining were in a decadent
state due to the effects of birch dieback and bronze birch borer attack (Nash,
Duda, and Gray, 1951).

Much research concerning the bronze birch borer and dieback in the 1940’s
and 1950’s was directed at determining whether or not the beetle was causing
the death of the trees. This question was hardly a new one then. Chittendon
(1898) says ‘‘as in the case of injury ascribed to the two-lined chestnut borer
[Agrilus bilineatus Weber] there is still a certain question of doubt as to whether
or not this birch borer is really the primary cause of the death of the trees.
Our correspondent is of the opinion that the injury in Buffalo is due primarily
to the attack of this borer, since it has been observed attacking vigorous trees. ’’
The work of Anderson (1944), Nash et. al. (1951), and Barter (1957) shows, how-
ever, that the insect played.a secondary role, usually killing those trees al-
ready afflicted with birch dieback. Whether or not birch dieback was the pre-
disposing factor in earlier ravages of the bronze birch borer, when it allegedly
killed mostly shade trees, is dependent upon what birch dieback really is. It
is evident that the shade trees in many cities were in a weakened condition in
the late 1800’s, perhaps because of droughtiness.

THE BRONZE POPLAR BORER. The habits of A. granulatus liragus adults
were first observed and recorded by Lintner (1883) before anything was pub-
lished about the habits of A. anxius, and of course well in advance of the estab-
lishment of the fact that the poplar feeder and birch feeder were not one and the
same. Chittendon (1898) says, ‘‘a very singular thing in connection with the
occurrence of this borer in birch is that in spite of frequent search, extending
over a period of two years, our correspondent has been unable to find this in-
sect attacking any other tree than birch--a remarkable condition of affairs when
we consider the numbers of observations by careful observers of its occurrence
on poplar and willow. ”’

Smith (1949a) discovered that the A. anxius (sensu Fisher, 1928) reared
from Betula spp. had 22 chromosomes, while those reared from Populus spp.
had only 20. Subsequently, Barter and Brown (1949) named the poplar feeding
form A. liragus. Although A. granulatus liragus has habits nearly identical
to those of anxius (host specificity excepted) it has never been accorded any
great significance as a pest. Like anxius the bronze poplar borer attacks and
kills only trees in weakened condition (Barter, 1965). because poplars are
used less frequently as ornamentals than birches, g. liragus is not as impor-
tant an ornamental pest as anxius. Furthermore, no large scale maladies on
the order of birch dieback occur in forests of native North American Populus
spp. so that the bronze poplar borer has not been associated with forest devas-
tation.

THE ASPEN ROOT GIRDLER. Harrison (1959) discovered what he thought to be
a new species of Agrilus feeding on the roots of aspen suckers (Populus tremu-

Carlson and Knight: Agrilus beetles 3

loides and P. grandidentata). The establishment of a chromosome number of
22 for this species seemed to indicate that it was more closely allied to anxius
than to g. liragus (Nord, 1962). Later, Nord, Knight and Vogt (1965) discovered
that the proper name for the root boring form is A. horni, a name which Fisher
(1928) and Barter and Brown (1949) thought to be a synonym of anxius. Because
of their uncertainty, however, they did not synonymize horni with anxius.

A. horni has attained a modest reputation as a pest. Nord (1962) says that
it caused severe damage to hybrid aspen out-plantings which were part of gen-
etics experiments conducted by the Institute of Paper Chemistry in Wisconsin
and Michigan. In 1964, the trees in some of the affected plots were destroyed
and the tests were reinitiated. However, in 1964 they were still having trouble
controlling horni! This, no doubt, was due to the inappropriateness of the con-
trol techniques used, rather than to any factor in the beetle’s ecology which
makes it difficult to control. Moreover, because of its ecological preferences,
it is not likely that this insect will ever cause great injury to well-stocked nat-
ural stands of aspen suckers that are growing on areas of moderate or good site
quality.

THE ALDER-BIRCH BORER. Fisher (1928) was uncertain of the status of A.
pensus. He says ‘‘this species may only represent a color variety of anxius,
but since nothing is known of its habits, it is best to retain it for the present,
at least, as a valid species.’’ Even though he was unsure of the distinction be-
tween pensus and anxius, he took the liberty of describing betulae, a synonym
of pensus, from 3 specimens. Although there is really little question that pen-
sus is actually a distinct species, its biology remained almost unknown prior to
the studies here described. The ecology of pensus makes it rather innocuous,
and it has attracted little attention. However, it appears to be an important fac-
tor in the ecology of alders (Alnus spp.), which are pecan ee as wildlife cover
in their lakeside and streamside habitat.

TAXONOMY
The Genus Agrilus

SIZE AND DISTRIBUTION . Agrilus beetles are thought to feed only on angio-
sperms (Fisher, 1928). Confinement to angiosperms could indicate that the
genus arose in the Cretaceous or later, which seems to allow ample time for
it to have attained the vast size and worldwide distribution it presently has.
Schaefer (1949) said the genus included 2500 described species, and Obenberger
(1957) claimed that in his collection alone, there were over 3000 described and
undescribed forms. Obenberger (1959) believed that there are over 4000 species
of Agrilus.

The largest number of species occur in tropical or subtropical areas, but
Agrilus is well-represented in temperate regions too. Fisher (1928) recog-
nized 125 described forms from the North American fauna (North of Mexico),
and a considerable number have been described since, mostly from the south-

|
Personal communication, 1964; D. Einspahr, Institute of Paper Chemistry, Rheinlander,
Wisconsin.

4 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

western United States.

SUBDIVISIONS. Obenberger (1957, 1959), with some rather specious rea-
soning, tried to show that there is no way to divide Agrilus into acceptable sub-
genera. He was, however, willing to admit that there are some recognizable
Species groups.

The economically important Eurasian ‘‘species’’! Agrilus viridis L., its
many destructive ‘‘host races, ’’ and the lesser known North American A. po-
litus Say together represent such a species group which is extremely difficult
taxonomically. Both A. politus and A. arcuatus Say are classified by Fisher
(1928) as being composed of several sympatric ‘“subspecies, ’’ but each actually
represents a multi-species ‘‘complex. ”’

Frost (1912) recognized some of these ‘‘sibling species’? problems, and
presented isa aes for their solution. He concluded by saying ‘‘the studies
in the otiosus’? and anxius groups have been so far, rather dissappointing, due
to the difficulty of getting series of both sexes. The only species that is at all
abundant in this locality is otiosus, taken on oak leaves. The olivaceous vari-
ety of acutipennis has been encountered quite often on oak, and bilineatus occurs
in favorable places on oak sprouts, but in general the species turn up singly or
in pairs, with aggravating slowness. Several very interesting problems are
suggested by the material in hand, and more specimens from widely separated
areas may present a solution.’’ In many groups of Agrilus, a similar situation
still exists.

The Agrilus anxius Group

DEFINITION. Presumably broadening the limits of the Agrilus anxius
group sensu Frost (1912), it is here defined as those species in which the adults
have the carina of the pygidium projecting as a spine, (figs. 2, 3), and (2) have
a sexual dimorphism in the structure of the last abdominal segment. The mar-
ginal sclerite? of the hind segment is smooth and in males lies in a plane near-
ly confluent with that formed by the sternite (fig. 33). In females the marginal
sclerite is reflexed or beveled with respect to the sternite (fig. 32). In most
of the species the first and/or second abdominal sternites are also sexually di-
morphic. In males these sternites are often grooved medially (fig. 1), while fe-
males never have such a groove.

NOMENCLATURE Nord et al. (1965) have referred to the ‘‘Agrilus anxius
complex.’’ The term complex carries a connotation of propinquity that is in-
appropriate for a group that includes anxius, g. liragus, and horni, and we,

anally includes a number of sympatric species (see, for example, Basurmanova, 1998).
2 eae
Agrilus otiosus Say.

“Those parts of the tergites termed the ‘‘vertical portions of the segments” by Fisher
(1928), but they are not vertical on the hind segments and the term marginal sclerite
seems to be a preferable alternative.

Carlson and Knight: Agrilus beetles D

therefore, will not apply it here. The Agrilus anxius group is comparable to
what in many other families of insects would be recognized as a genus. Were
the anxius group to be delineated from other Agrilus as either a subgenus or
genus, either of the Agrilus synonyms Teres Harris (1829) or Uragrilus Seme-
nov (1935) (=Epinagrilus Stepanov, 1954) could be applied. Because the limits
of the anxius group are not ascertainable from the literature, it would be un-
wise to use a formal name for it at this time.

MORPHOLOGIC ANALYSIS . The taxonomic analysis of anxius group species
is based primarily on adult morphology, but important information is also fur-
nished by other developmental stages. The interpretation of morphologic char-
acters is strongly influenced by studies of adult behavior which are discussed
later, and some findings from studies of adult morphology are deferred to the
Evolution section.

Adults. The analysis of adult morphology was based primarily on collec-
tions made in Michigan (Iron, Gogebic, Benzie, and Manistee Counties) and
Wisconsin (Vilas Co.). Material in the collections of the University of Michigan
Museum of Zoology (Ann Arbor, Michigan) the Entomology Research Institute
of Canada (Ottawa, Ontario), Harvard University, Museum of Comparative
Zoology (Cambridge, Massachusetts), the Philadelphia Academy of Sciences
(Philadelphia, Pennsylvania) and the U. S. National Museum (Washington, D.C.)
was also examined. The following types were studied: Agrilus betulae Fisher,
Agrilus fulminans Fisher, Agrilus populi Fisher, Agrilus is horni Kerremans
(=blanchardi Horn, name preoccupied), A Agrilus nevadensis Horn, Agrilus
niveiventris Horn, Agrilus pensus Horn, and Agrilus gravis Leconte.

On the basis of the study of the types and other material the following no-
menclatural changes (including those made by Carlson [1969]) were called for
and will be followed here:

AGRILUS BROWNI, new species

Agrilus browni, new species
Agrilus lateralis, sensu Fisher (1928), in part.

Fig. 1. Ventral aspect of Agrilus horni “, show-
ing the median groove on abdominal sternites 1 and 2.

6 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

AGRILUS GRANULATUS Say

Agrilus granulatus granulatus Say
Buprestis granulata Say lata Say (1823)
Agrilus grant granulatus populi Fisher; status assigned by Ss (1969)
- Agrilus populi Fisher (1928)
Agrilus Erichoca tpae Chamberlin (1929)
Agrilus granulatus liragus Barter and Brown; status assigned by Carlson (1969)
Agrilus liragus Barter and Brown (1949)

AGRILUS PENSUS Horn

Agrilus pensus Horn (1891)
Agrilus betulae Fisher (1928); synonymy designated by Carlson (1969)

AGRILUS QUADRIGUTTATUS Gory

Agrilus quadriguttatus quadriguttatus Gory
Agrilus quadriguttatus Dejean (1833, 1836), nomen nudum

Agrilus quadriguttatus Gory (1841)
Agrilus quadriguttatus niveiventris Horn; status assigned by Carlson (1969)
Agrilus niveiventris Horn (1891)
Agrilus nevadensis Horn (1891); synonymy designated by Carlson (1969)
Agrilus quadriguttatus fulminans Fisher; status assigned by Carlson (1969)
Agrilus fulminans Fisher (1928)

AGRILUS RUFICOLLIS Fabricius

Agrilus ruficollis Fabricius
Buprestis ruficollis Fabricius (1787)
Agrilus lateralis Say (1823), new synonymy
Agrilus sayi Saunders (1871), name unnecessary (lateralis not preoccupied)

Fisher (1922) reared an Agrilus sp. from bayberry (Myrica pensylvanica,
= carolinensis) at Lyme, COBER CEI for which he used the name A. lateralis
Say. The original description’ of lateralis (Say, 1823), however, fits A. rufi-
collis Fabricius more closely than the species that was reared from Myrica.
Forthat reason and because of the range limits apparent for it, the species
which Fisher reared from Myrica is here given the name Agrilus brownin. sp.

Say (1823) says the head of lateralis has a profoundly impressed line, which
certainly is characteristic of ruficollis. A. browni has the front only weakly
impressed. Say also notes that the length of lateralis is1/5 inch (about 5.1 mm),
which is not exceptionally short for ruficollis, a smaller and more variable
species than browni. On the other hand, specimens of browni as short as 5.1
mm must be extremely rare, as we have seen none approaching that size.

I enordine to Fisher (1928) the lateralis type was lost, but, presumably, it was destroyed,
as is the case for the majority of Say’s specimens.

Carlson and Knight: Agrilus beetles 7

The only bonafide records for browni are from places very remote from
the lateralis type locality. Say (1823) described lateralis from a specimen col-
lected in Missouri, which then included the vast region west of the Mississippi
known as the Louisiana Purchase, excluding what is now the state of Louisiana.
Knull (1922) found browni in abundance on sweet ferns (Myrica asplenifolia, =
Comptonia peregrina) in Pennsylvania, but no reliable records are known to
us for any area west of that state. In spite of the examination of large amounts
of sweet fern by P. C. Kennedy! and a modest amount of searching on that host
plant by us, browni has not been encountered in Michigan. The strength of all

the evidence leads us to suppress lateralis as a synonym of ruficollis.

Fisher (1928) provides a good description of browni (under the name later-
alis), and his ‘‘lateralis’’ neotype? is here designated as the type of browni.
The species is named for Mr. William J. Brown?, who suggested that Fisher
(1922, 1928)had misapplied the name lateralis.

Fisher (1928) knew A. nevadensis Horn only from the type, and that speci-
men appears to have been deformed during pupal development. Fisher (1928)
says its scutellum is not carinate, but the carina appears to be present, al-
though it is positioned abnormally on the deformed scutellum. The somewhat
prolonged elytra have more sharply pointed apices and more coarsely rugulose
surfaces than typical q. niveiventris, but there are no other structural differ-
ences. The pigmentation of Some specimens of q. niveiventris from southern
California is as dark as that of the A. nevadensis type, which was collected in
Nevada. ee

The following sections will be devoted primarily to anxius, pensus, horni,
and g. liragus. The discussion of other members of the anxius group is de-
ferred to a later section on evolution.

Since Gory (1841) described anxius, other forms have been named, two of
which are synonyms, while another was mistaken for a synonym. Agrilus
gravis Leconte and A. torpidus Leconte were described in 1859, but Fisher (1928)
placed them as synonyms of anxius. Agrilus blanchardi Horn (not Saunders) and
A. pensus Horn were described in 1891. Because the former was a primary
homonym, it was renamed A. horni by Kerremans (1900). Fisher (1928) thought
that horni was a synonym of anxius. Knowing of no material besides the type
Which could be identified as horni, he believed that it was merely an abnormal
specimen. Being somewhat uncertain, however, he retained horni as valid.

When Barter and Brown (1949) established that a poplar-feeding form and
a birch-feeding form had been confused under the name anxius, they decided

a He communication, 1967; P. C. Kennedy, North Central Forest Experiment Sta-
tion, U. S. Forest Service, East Lansing, Michigan.

en to Fisher (1928), this specimen is a male, which he collected on bayberry at
Lyme, Connecticut and deposited in the U. S. National Museum collection.

Uehncapliony Research Institute, Canada Department of Agriculture, Ottawa, Ontario.

8 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

to retain’ that name for the birch-feeding form. They suppressed torpidus

as a synonym of anxius, but made no definite decision regarding the name gra-
vis. The gravis type is a female that Barter and Brown (1949) could not identi-
fy reliably. The name liragus was, therefore, given to the poplar feeding form
with the distinct possibility of it later proving to be a synonym of gravis. There
have been doubts that this was the proper thing to do. For example, the appli-
cation of gravis by Blanchard (1889) for the species collected on poplars (not
‘“sprouts’’2) is cited by Nash et al. (1951) as a possible indication that gravis —
was the proper name for the poplar feeder.

_ Although the gravis type lacks the coppery sheen often found on the head and
pronotum of anxius specimens, its absence is not a good indication of the iden-
tity of the type. Vogt? pointed out that specimens of anxius from some parts
of the Great Lakes Region are often colored more like g. liragus than are spec-
imens from more eastern parts of the United States. Indeed, the coppery re-
flections are often very indistinct in specimens from the Upper Peninsula of
Michigan, which lies within the type locality recorded by Leconte (1859) as the
‘*Lake Superior Region. ’’

By relying on characteristics other than coloration and working with spec-
imens collected in the Upper Peninsula of Michigan, we found that Agrilus gra-
vis Leconte is really a synonym of anxius Gory, and therefore g. liragus Bar-
ter and Brown is to be retained as valid The name gravis, as used in a num-
ber of papers in the late 1800’s, refers to either anxius or g. liragus; assign-
ment to one or the other depends upon the host designated, if any. References
using the other anxius synonym, torpidus Leconte, are to be treated similarly,
except for those which refer to willow as a host. The latter (e.g. Cook, 1890;
Davis, 1891, 1892) refer to A. criddlei Frost, which makes galls on willows.
The host of criddlei was recently established through rearing by Wong and Mc-
Leod (1965).

Barter and Brown (1949) concluded that anxius and g. liragus are separable
only by reliance on male genitalia, coloration of the head and pronotum, and
host plant identity. We cannot concur with them fully. As previously indica-
ted, anxius specimens from some areas do not have very distinct coppery re- |
flections on the pronotum, and this iridescence also seems to fade quite rapid-
ly in Some museum specimens. Wetting the specimens with various relaxing
agents is particularly destructive to the iridescence of these beetles. On the
other hand, there are structural differences, besides male genitalic ones which,
in combination, can be used for separating anxius and g. liragus. One of these,
a difference in the form of the tarsal claws, was not noted by Barter and Brown
(1949). The membraneous attachment of the tarsal claws of g. liragus appears
to be either more flexible or looser than that of anxius. Consequently, the g.

liragus tarsal claws are often clearly separated from the unguifer, making the

Aathey, indicated the location of the anxius type as being unknown.
“Blanchard thought what he had found on poplar sprouts was anxius.

*Dersonal commununication, 1966; G. B. Vogt, U. S. Dept. Agr., A.R.S., Entomology
Res. Div., Washington, D. C.

Carlson and Knight: Agrilus beetles | 9

latter clearly visible. The claws of anxius are attached so that there is very
little space between their bases and the unguifer. The unguifer, is therefore,
more difficult to see. In addition, the claws of anxius are usually rather dis-
tinctly swollen at their bases, and tend to obscure their attaching membrane
which stretches across the opening in the last tarsomere, while in g. liragus the
basal parts of the claws are usually not strongly swollen, and the membrane in
the tarsomere opening is usually pti visible.

The females of anxius and g. liragus differ in the shape of the projecting
Spine on the pygidium, im, but the males do not. In g. liragus females the spine
usually projects strongly (fig. 2) and is rather cylindrical, while in anxius fe-
males it is usually shorter and more blunt (fig. 3), and its tip is often broad
and elyptical in end view.

A useful characteristic mentioned by Barter and Brown(1949), is the ten-
dency for the elytral tips of g. liragus to be acute and strongly serrate (fig.

Fig. 2. Dorsal surface of the pygidium
of an Agrilus granulatus liragus 2 (X40).

Fig. 3. Dorsal surface of the pygidium
of an Agrilus anxius ? (X40).

10 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

A B C

Fig. 4. Elytra of A. granulatus liragus o (A), A. anxius ¢ (B), and A. pensus ¢ (C).

4a), while those of anxius are more broadly rounded and not as strongly
toothed (fig. 4b). By using both the coloration differences and the somewhat
obscure and somewhat variable structural differences of the pygidial spine,
tarsal claws, and elytra it is usually possible to distinguish the females of an-
xius and g. liragus. iets

The most reliable characters for separating males of anxius and g. lira-
gus are those of the aedeagus. The parameres of anxius have broad ventral
inner margins, and these margins together form a trough which is broad and
shallow (fig. 5). Ventrally, the inner margins of the g. liragus parameres are
usually narrow, together forming a trough which is narrow but deep in compar-
ison with that of anxius (fig. 6).

Nord et al. (1965) separate horni from anxius and g. liragus by utilizing
characteristics of both the male and female genitalia. The ovipositor of horni
(relative to elytron length) is significantly shorter than that of g. liragus or
anxius. Nord et al. also pointed out that the average ovipositor length is less
in g. liragus than in anxius, and in some cases the minimum length for g. lira-
gus approaches the horni maximum. However, they found no overlap in ovi-
positor length between g. liragus and horni.

The male genitalia of these species correspond in length to those of the fe-
males. In addition to being shorter, the aedeagus of horni differs from those
of anxius and g. liragus in having parameres which are less dialated posterioral-
ly, their margins being subparallel (fig. 7). Nord et al. (1965) say that ven-
trally and subapically the inner margins of horni parameres are arcuate to
subarcuate, while those of anxius and g. liragus are angulate to subangulate.

Carlson and Knight: Agrilus beetles 11

Figs. 5-8. Ventral sides of the aedeagi of Agrilus anxius (5), A. granulatus
liragus (6), A. horni (7), and A. pensus (8).

The ventral trough formed by the parameres tends to be intermediate in form
to the troughs of anxius and g. liragus, which are quite different from each
other.

Despite the confusion which has occurred regarding the identity of horni,
it is really quite distinct from either anxius or g. liragus. Besides the geni-
talic differences already noted, horni differs significantly from anxius and g.
liragus in other ways. Most significant of all, perhaps, is the moderately
dense pubescence on the marginal sclerites of the abdomen. As Horn (1891)
points out, this pubescence appears all the more distinct because of the white
pruinosity which is associated with it. Together, the pruninosity and white
pubescence tend to form a white band which is very distinct on newly-collected
horni specimens, but also quite noticeable on material that has been preserved
for many years. The pubescence is usually dense enough to obscure the sculp-
turing of all the marginal sclerites, except for the last one. These marginal
sclerites are usually less pubescent in anxius and g. liragus, and the pubescence
is usually not dense enough to obscure the sculpturing of the marginal sclerites.
Moreover, the marginal sclerite of the second abdominal segment is either
glabrous or noticeably less pubescent than those of the first and third segments
in anxius and g. liragus.

There are also distinct differences in the structure of the elytra. The tex-
ture of horni elytra is very even, the disc being comparatively smooth, finely
imbricate, and usually lacking a distinct longitudinal costa (fig. 9). The disc
is usually more coarsely imbricate in anxius and g. liragus, and often has a

12 Contrib. Amer. Ent. Inst. vol. 4, no. 3, 1969

Pig... 9.) Agrilue horni ¢,

longitudinal costa and a rather distinct depression just mesad of the costa and
between the disc’s basal third and apical half. Frequently in g. liragus, and
sometimes also in anxius (see o’, fig. 39), this depression is pruinose, the
pruinosity forming a spot similar those on the elytra of g. granulatus, but less
distinct. Upon close examination, the parts of the depression where the spot
tends to form can usually be seen to have a small group of recumbent hairs
which are alingned differently and sometimes longer than the surrounding ones.
The pruinosity usually disappears in preserved specimens, but the faint spot
formed by the unaligned hairs can often be detected.

The recumbent pubescence on the elytra of horni is distributed rather e-
venly over the surface and is also rather short and uniform in length. That of
anxius and g. liragus is usually not uniform in distribution or length, and many
of the hairs are longer than those of horni elytra.

The structure of the head is also somewhat different in horni. The front
is almost always rather flattened in profile. Sometimes in anxius and even
less frequently in g. liragus, the front is somewhat flattened in din profile, but
in these species it usually bulges, presenting a rounded profile.

The form of Agrilus pensus is intermediate between that of A. anxius and
A. quadriguttatus Gory, the latter being a willow feeding species. Barter and
Brown (1949) say that the male genitalia of pensus are structured like those of
g. liragus and g. populi, but this is not true. The pensus aedeagus (fig. 8) is
somewhat variable, but more like that of anxius than that of g. liragus or g.
populi. Barter and Brown also say that pensus lacks pubescent and prui-
nose spots on the elytra, but while they are not found on specimens from Nova
Scotia and New Brunswick they are quite apparent on some specimens from
Michigan and are also found on the type of the pensus synonym betulae Fisher.
As in A. quadriguttatus and subspecies the subapical maculation of n of the elytra
is likely to be spread out along the elytral suture instead of concentrated in
a small spot. Also as in A. quadriguttatus the elytra are likely to be angulate
at the apex (fig. 4c), forming a broad notch, instead of acuminate as in A.
eee s and subspecies, which have a narrower notch between the apices.

Eggs. The eggs of pensus (Engel, 1968), anxius, g. liragus, and horni
are creamy white immediately after they are deposited. They are flattened and

Carlson and Knight: Agrilus beetles 13

oval in outline. There seem to be no interspecific differences in egg size,
those of horni being about 1.2 mm long by 0.8 mm wide, according to Nord
(1962).

The eggs of anxius (and presumably g. liragus) are coated with a semi-
transparent whitish substance immediately following deposition (Barter, 1957).
The substance apparently acts as a cement and as a protectant for the soft egg
shell. There is no evidence that horni applies such a substance to its eggs.
Instead, the chorion of horni eggs appears to toughen as the larvae develop,
and as a result they turn dark brown to black. The eggs of anxius and g. lira-
gus darken like this only when parasitized. They develop grayish- -black streaks
and then become completely black prior to emergence of the parasite adult (cf
Barter, 1957, 1965). This led Nord et al. (1965) to believe that the darkening
of horni eggs may have been associated with parasitism, but this is not the
case. |

Larvae. Benoit (1965) presents a complete description of g. liragus larvae,
and gives the differences between these and the larvae of anxius. He says the
larvae of these species are about 30 to 40 mm long and 2 to 3.5 mm wide .

The body is white, except for the anal forceps and other minor sclerotized por-
tions, which are ferruginous. The prothorax, which is wider than the rest of
the body, has a median I-shaped sclerotized line on the tergum. A similar
structure, which is only vaguely sclerotized, is located on the prosternum
(Barter, 1957). The segments posterior to the prothorax are all about equal

in width.

Benoit (1965) says C-shaped spiracles are located on abdominal segments
one through eight, and similar, but somewhat larger ones are found dorsolat-
erally on the mesothorax. Actually, the sclerotized C-shaped structure is
part of the spiracular valve, and the spiracular opening itself is located inter-
nal to the transparent valve.

Benoit (1965) states that the labial palpi are absent, but this is not true.
Apparently he missed them because they were out of focus when his microscope
was focused on the rest of labium (cf fig. 10 and his fig. 1B).

Because of the nature of his samples, Benoit’s morphologic comparison of
the larvae of anxius and g. liragus is of questionable value. He obtained his
material from two localities, one for each species. This could hardly give one
an adequate sample of intraspecific variation.

Nevertheless, the species differences in the coloration of the anal forceps
which Benoit has listed do appear to hold for larvae collected in Michigan as
well as for those from Quebec. The anal forceps of anxius are described by
Benoit (1965) as being reddish, with the line of sclerotization extending anteri-
orally beyond the anal opening. He says the forcepsof g. liragus are more
brownish, with the sclerotization ending abruptly just posterior to the anal o-
pening. The larvae of pensus and horni tend to be intermediate to those of
anxius and g. liragus in both the structure and coloration of the anal forceps.

~The characters Benoit (1965) used were worked out by Alexseev (1960).
Among them was the location of a short macro-seta (‘‘un pied petite’’) located
in relation to a small circular depression or dimple in the cuticle. The dim-

1 :
Benoit (1965) does not specify the developmental stage of the larvae from which his mea-
surements were taken. ,

14 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Fig. 10. Labium of Agrilus granulatus liragus lar-
va (X234). Note that the right palp is completely out of

focus as with both in fig. 1B of Benoit (1965).

ples which are supposed to be associated with the pedalis structuralis (thoracic
leg rudiments) are not evident on living larvae, and do not seem to be present
with any regularity on specimens fixed in ‘‘XAAD’’ or by boiling. Consequent-
ly, the dimple may be the result of shrinking or swelling of the cuticle or as-
sociated tissues which may be related to the method of fixation and preserva-
tion. Benoit (1965) killed and preserved his larvae in 70 percent alcohol.
There also seems to be the possibility that the absence of the depressions may
have been related to the stage of larval development. Most of the larvae we
examined had assumed the bent-over position in the pupal chamber. Benoit
failed to mention the developmental state of the larvae he examined.

Benoit (1965) says there are also other marked differences between the
larvae of anxius and g. liragus. As these require higher magnifications or
more preparation, we did not examine Michigan material for these characters.

Pupae. There is nothing particularly unusual about the pupae of these spec-
ies. All are creamy-white at first, and gradually assume the adult coloration.
The eyes are the first parts to darken, and the elytra darken last. It appears
to be during the period of pupal development that much of the adult structural
variation arises, and it may very well be as much due to variation in the con-
ditions under which the pupae develop as it is to genotypic variation.

GENERAL BIOLOGIES OF AGRILUS ANXIUS,
A. PENSUS, A. HORNI, AND
A. GRANULATUS LIRAGUS

Hosts and Ranges
THE BRONZE BIRCH BORER. A. anxius appears to be host specific to birches,

Betula spp. Fisher (1928) lists the following birches as hosts for anxius:
yellow birch, Betula lutea; paper birch, B. papyrifera; European white birch,

Carlson and Knight: Agrilus beetles 15

B. alba (and varieties); gray birch, B. populifolia; and sweet birch, B. lenta.
Barter and Brown (1949) also list B. fontinalis as a host for anxius. Hitchcock,
et al. (1964) rank this birch as a variety of the western paper birch, B. oe
dentalis, which is closely allied to B. papyrifera. Harlow and Harrar (1958),
in fact, call the western paper birch B. papyrifera var. commutata.

“Although it has been reared from numerous hosts, the principal hosts for
anxius are the native paper birches, B. papyrifera and B. occidentalis. A.
anxius is, then, primarily confined to northern forest types, but its range ex-
tends somewhat south of the range of B. papyrifera in the eastern U. S., where
it is found even at low altitudes as far south as Washington, D. C. It ranges
west to Oregon and British Columbia.

THE BRONZE POPLAR BORER. A. granulatus 1 iragus has been reared from
several Populus species, although ‘its principal host is trembling aspen, Pop-
ulus tremuloides. In addition, Barter (1965) lists the following as hosts for
g. liragus: bigtooth aspen, P. grandidentata; eastern cottonwood, P. deltoides;
balsam poplar, P. balsamifera ssp. balsamifera; and the black cottonwood, ¥%
balsamifera ssp. p riehooarpe (Torr. & Gray) Brayshaw (1965).

The known range of g. liragus is very similar to that of anxius, but g. lir-
agus does not occur as far south as anxius does in the eastern U. S. The
southern edge of the g. liragus range seems to comply more closely with the
range of P. tremuloides than that of anxius does with the range of B. papyrifera.
Just how far north either g. liragus or anxius occur is not known, ‘but both
have been taken as far north as Edmonton, Alberta. The former, at least,
should occur in interior Alaska.

THE ASPEN ROOT GIRDLER. Trembling aspen is the principal horni host,
but bigtooth aspen and balsam poplar are also infested. Infestation levels on
balsam poplar are sometimes quite high.

When seed stock of various species of poplars are transplanted in experi-
mental orchard-like plots, horni can be particularly troublesome. Nord et al.
(1965) found that under these conditions, horni readily infests the Eurasian
species Populus alba and P. tremula. He found that various hybrid aspens
were also infested. In nature, however, horni is probably confined mostly to
species which reproduce by root suckering (i. e. aspens and balsam poplars).

Although horni has not yet been found as far west as g. liragus, it is prob-
ably not because the former has amore limited distribution, but is more likely
a reflection of the fact that horni is not collected as readily as is g. liragus.
There are specimens inthe U. S. National Museum from Massachusetts to
Arizona, and as far north as Aweme, Manitoba.

THE ALDER-BIRCH BORER. A. pensus has been reared from Alnus rugosa
and collected on the foliage of this shrub. It has also been found in abundance
on the foliage of green alder, Alnus crispa according to Barter and Brown,
(1949). Fisher (1928) described betulae, a synonym of pensus, from speci-
mens reared from river birch (Betula nigra) and Knull (1930) reared a large
series from that host. The records of Blanchard (1889) for Agrilus granula-
tus collected on alders refer to pensus | , not to quadriguttatus as Fisher (1928)
suggests. Records for pensus on Ostrya virginiana are of questionable sig-
nificance. One specimen collected by C. A. Frost (Monmouth, Maine; June
23, 1910) is labeled ‘‘ironwood’’ and three by Geo. M. Greene (Philadelphia,

1
Horn (1891) described pensus from the series taken by Blanchard (1889).

16 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Pennsylvania; May 24, 1900) are labeled ‘‘hophornbean.’’! Not only is it rath-
er doubtful that pensus infests Ostrya, but it is entirely possible that alders
or young birches were mistaken for Ostrya. Ironwood, at least, is an unlike-
ly host because its ecology is so very different from that of alders and river
birch. A. pensus has been collected as far west as Aweme, in southern Man-
itoba, and in the East occurs as far north as mid-New Brunswick (Barter and
Brown, 1949). Manee (1913) recorded Agrilus granulatus on ‘‘black alder’’

at Southern Pines, North Carolina. As it was collected with Eupristocerus
cogitans Weber (an agriline species that infests alders), the record should be
considered to refer to Agrilus pensus.

Life Cycles

A. anxius and A. granulatus liragus have a two-year life cycle under some
conditions, but a one-year life cycle when other conditions prevail. A. horni
and A. pensus, however, seem to have obligatory two-year life cycles.

ANXIUS AND G. LIRAGUS. When the eggs of anxius and g. liragus are de-
posited on hosts that are dying, severely injured trees or blowdowns, for ex-
ample, the larvae often construct the pupal chamber during the same season
that the eggs were laid. On the other hand, when oviposition occurs on more
vigorous hosts, a two year life cycle results (Balch and Prebble, 1940; Ander-
son, 1942; Barter, 1957, 1965). Nonetheless, the larvae of these species
cannot survive in healthy trees, and successful larval development is always
dependent upon the host being in a weakened condition because of drought, re-
peated unsuccessful borer attacks, or other injuries (Barter, 1957, 1965).

PENSUS. The habits of A. pensus have not been studied sufficiently to es-
tablish the usual condition of its host at the time of oviposition. We have never
seen larval galleries that would indicate a life cycle of less than two years
for pensus. Furthermore, the larvae appear to overwinter primarily in two
stages. Mature larvae, of course, overwinter in the pupal chamber, and
early instar larvae are found overwintering deep within the xylem. Blanchard
(1889) says pensus ‘‘breeds in and frequents stems of partly dead alders, but
does not appear on the foliage very much.’’ Knull (1950) says it breeds in
‘‘unhealthy’’ river birches, but mentions nothing about the size of the trees in-
fested. Fisher (1928) says the pupal cells are common in the stems of young
river birches that have died, and we have similarly found that the adult emer-
gence holes are common on dead alders. However, numerous speckled alders
were examined for the presence of pensus larvae, and those which harbor im-
mature larvae are invariably alive, and many of those infested with mature
larvae or later stages have not succumbed completely. Because the larvae al-
ways seem to require more than one year to complete development, the adults
probably never oviposit on hosts imminently near death.

Only one attempt was made to attract pensus with cut stems of alders.
Although there was a pensus infestation in the area, none were attracted. Ovi-
position on such material would be detrimental for this species, because early
death would be inevitable for the larvae. The bark of Alnus rugosa is so thin
that the phloem would become moribund and dessicated much too quickly to

Noseinens in U. S. National Museum, Washington, D. C.

Carlson and Knight: Agrilus beetles 17

permit larval development. However, it seems probable that even large, de-
cisively injured or felled river birches are not attractive to pensus.

HORNI. Nord (1962) found no evidence of anything less than a two-year
life cycle for horni, and those findings are supported by our observations. By
examining suckers containing mature larvae it is often possible to establish
when the first instar gallery was formed. First instar larvae feed in the phloem!
while moving down the stem toward the root, but in no instance was a first in-
star gallery located in the phloem of suckers infested with mature larvae. When
it was not obliterated by the larva when it re-entered the stem, the dark spot
indicating the position of the first instar gallery was invariably found between
the outer two rings of xylem.

Moreover, out of 60 larvae introduced into suckers of various sizes, 5 lar-
vae completed their development during the second summer. One larva ma-
tured in the laboratory during the third summer, but no larvae matured during
the first season.

Nord (1962) found that the total length of horni galleries varied between 40
and 90 inches. The maximum length of the galleries of anxius and g. liragus
which mature in two years is about 52 inches. The galleries of anxius or g.
liragus larvae completing their development in one year are usually less than
half that long (cf Barter, 1957; Anderson, 1944). Thus the evidence is over-
whelmingly against the occurrence of a one-year life cycle for horni. Two
years are usually required for horni development to be completed, but occa-
Sionally it may take three years.

Developmental Patterns

HATCHING OF EGGS. Barter (1957, 1965) found that incubation of anxius
and g. liragus eggs requires about two weeks. The same is true for horni,
and is probably generally true of many Agrilus. When embryonic development
is complete, the first larval instar bores out of the egg and into the bark of the
host, leaving the egg shell packed with frass.

LARVAL FEEDING PERIOD. Although larval behavior differs among Agrilus
species, certain generalities can be made. As with most Buprestidae, the lar-
vae never bore exit holes for their frass, but instead pack it tightly into the
gallery behind them.

Heering (1956) gives a very detailed account of the larval behavior of Agri-
lus viridis. Some aspects of its behavior seem very similar to those of anxius
or g. liragus, and other aspects seem typical for Agrilus in general. The viri-
dis larvae are pressed tightly into the gallery. According to Heering this allows
the microscopic spinules on the lateral muscular ridges to grip the sides of the
gallery. The larvae move forward by constricting and stretching the segments
telescopically. Movement of the parts of the body besides the head and anal for-
ceps is entirely restricted by the sides of the gallery.

All the tissues macerated by the larvae pass through the digestive tract.

As the frass passes out the anus, it is clasped between the anal forceps. With
these the larvae pack the frass tightly into the gallery behind them (Heering,

lord et al. (1965) said that during the first stadium the larvae feed in the cortex, but
this is the position of the gallery one year after it is formed.

18 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

1956). It would seem necessary that the frass be packed at least enough to
make its density as great as that of the tissues on which the larvae feed.

The Bronze Birch and Bronze Poplar Borers. Barter (1957, 1965) says
that after they. bore out of the egg, the larvae move directly through the bark
to the cambial'’ region, but Anderson (1944) found that some larvae bore as
much as 10 inches in the bark while passing obliquely to the cambial region.
These species mine chiefly in the phloem, but the xylem is also scored. An-
derson (1944) found that after reaching the cambial region, the larvae feed for
an average of 1.3, 3.9, 7.1, and 13.0 inches in each of the four larval stadia
preceding the molt to the fifth instar, after which the pupal chamber is formed.
This indicates that there is a correlation between larval size and the amount of
feeding done in any stadium. In the trees from which Anderson accumulated
these data the points at which molting occurred were marked by departures of
the gallery into the xylem. In hosts of a very decadent nature the larvae usu-
ally do not bore into the xylem before molting. On the other hand, in the more
vigorous hosts which can be successfully infested, there are likely to be more
xylem departures than molts. Apparently, these xylem departures are elic-
ited by the unfavorable conditions occurring when host vigor is comparatively
high.

Host condition also has a bearing on other aspects of gallery formation.

In severely weakened or very decadent hosts larval galleries usually show no
distinct pattern. Under these conditions, the larvae follow a course suggesting
that they feed on the freshest phloem they encounter (Barter, 1957). These
galleries can be fairly straight, but they usually meander without any pattern
in these kinds of hosts.

In more vigorous hosts a zig-zag or sinuate gallery pattern is the rule.
Under these conditions anxius makes a gallery which progresses in the direc-
tion tended by the phloem sieve elements or the xylem vessels, but weaves
back and forth across their grain forming a gallery with continually expanding
hyberbolic links (figs. 11, 16). Under similar conditions the successive links

Fig. 11. Larval ely of Agrilus anxius (in Betula papyrifera)
that has healed over, indicating that it was made under conditions

of high host resistance (from Anderson, 1944).

eine authors suggest that these insects bore in the cambium. The cambium is extreme-
ly thin (strictly defined, a single cell thick; cf Esau, 1953) and, therefore, it is inaccur-
ate to say that these insects bore in the cambium. Although the cambium may be con-
sumed as an insignificant part of the nutriment, food value is more likely to be derived
from its derivatives, particularly the phloem and perhaps the newly-formed xylem.

Carlson and Knight: Agrilus beetles 19

of the g. liragus gallery are much closer together, often lying adjacent to each
other (fig. 12). The attempt to explain these gallery patterns will be deferred
to the part of this section which deals with host resistance.

The Alder-Birch Borer. The larvae of A. pensus boring in the stems of
Alnus r rugosa iosa often form galleries without any regular pattern. Sometimes,
however, the gallery is sinuate like those of anxius, particularly those parts
of the pensus gallery which do not lie deeply in the wood. Xylem departures
are more frequent with pensus larvae, and sometimes they bore through the
wood from one Side of the stem to the other, a behavioral trait noted by Bar-
ter (1957) for anxius when it bores in branches.

According to Larsen (1901) anxius larvae make spiral galleries in branch-
es. This has never been reported for g. liragus, and pensus does not form
Spiral galleries, even when it bores in alder branches only about an inch in
diameter.

Only one case was encountered in which several pensus larvae had com-
pletely girdled the phloem! , and this was only during the latter stages of lar-
val development. Very little is known about the larval behavior of this spec-
ies, but it seems clear that if the host dies during the earlier stages of lar-
val development, they would be unable to complete their development (cf page
16). Perhaps for this reason, mass attack is not the habit of this insect.

The Aspen Root Girdler. Some aspects of A. horni development seem to
differ radically from what is characteristic of anxius, pensus, and g. liragus.
In horni and often in the others, eclosion from ‘the eg egg is followed by direct
entry into the phloem, but the first instar gallery is otherwise quite different
in horni. The mechanism triggering the first molt of horni larvae appears to

Fig. 12. Larval galleries of Agrilus granulatus (from area of in-
tergradation ofthe nominate subspecies with g. liragus [see p. 85],
but typical for the latter, at least). Some had healed over, indicat-
ing high host resistance (courtesy of R. E. Stevenson, Forest Biol-
ogist, Canada Dept. For. & Rural Dev. , Calgary, Alberta).

OP on ieke girdling may be more prevalent under some conditions than others. At Mall-
ard Creek in Iron Co., Michigan, a moderately large pensus population build-up occurred
one or two years after a stand of alders had been flooded by a beaver dam. In this area,
alders that had been completely girdled by pensus were not rare (personal communication,
1968; R. P. Engel, University of Michigan, Ann Arbor).

20) Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

be different, because the first instar galleries of this species are comparative-
ly long. Prior to molting the first instar larvae of anxius and g. liragus bore
an average of 1.3 inches after entering the phloem, although the gallery some-
times extends as much as 10 inches in the bark before it reaches the phloem
(Anderson, 1944). In suckers recently killed by horni what appears to be the
first instar gallery may extend out into a root for as much as 42 inches, and
the total length of the first instar gallery usually is greater than 12 inches.
Whether or not the first molt of the horni larvae occurs during the first sea-
son of development is unknown. No larvae have been found at this stage during
the winter or spring. Presumably, they have been overlooked because of their
small size (about 2 mm long in the first stadium).

Brooks (1914), presents an excellent discussion of the larval behavior of
Agrilus vittaticollis Randall, which probably is a distant relative of horni
(see fig. 44). He says that upon leaving the egg, the vittaticollis larva bores
in the ‘‘cambium’’ (meaning cambial region) progressing down the trunk and
into the root for a distance of 6 to 12 inches before boring into solid wood.

For most of its length the gallery in the phloem is straight, generally following
the grain of the wood. However, it sometimes zig-zags across the grain af-
ter leaving the egg, and invariably spirals once or twice in the phloem of the
root before entering the xylem.

The larvae of horni behave similarly, but proceed with the grain until they
enter the root. Just after reaching the root they are apt to make one or two
Spirals in the phloem, and often make others before they begin boring princi-
pally in the xylem. This behavior varies and may be dependent upon the size
of the root and the condition of the host.

Sometimes the horni larvae do not bore in the xylem until they turn back
toward the stem. In other instances they spiral in the xylem while still pro-
gressing away from the stem. Root diameter seems to be the factor deter-
mining where the larvae change the feeding direction. The turn is apparently
elicited when the larvae reach a point where the diameter of the root they oc-
cupy is so small that they completely destroy the xylem. In aspens this is apt
to occur where the root diameter is about 1/8 to 3/16 of an inch in diameter.
The larvae may derive the stimulus eliciting the turn toward the stem when
severing of the xylem results in different pressures on each side of the breach.
While transpiration should make the pressure on the proximal side negative,
any pressure in the distal part should be positive.

The turning point may be as much as 42 inches or more from the stem, but
if the roots are short the larva may reverse its feeding direction only 6 inches
from the stem. Instances in which the turning point is close to the stem are
rare because the roots of aspen suckers are usually rather long.

The phloem in the roots of young aspen suckers is much thinner than that
in aspen or birch stems suitable for infestation by g. liragus and anxius, re-
spectively. The difference relates to the age-dependent accruement of inac-
tive phloem. Although horni galleries lie principally within the xylem, it
could hardly be otherwise. It appears that they consume the phloem too, and
of all they ingest, the phloem probably provides most of what is assimilated.

After horni larvae re-enter the stem they sometimes bore a rather loose-
ly spiralled or almost straight gallery which departs from the cambial region.
Nord (1962), however, notes that the larvae always tunnel in the cambial re-
gion long enough to girdle the phloem. The root damage inflicted by horni
larvae rarely seems to cause the death of suckers, because foliar discolora-
tion occurs after the larvae return to the stem. Zahner and Debyle (1962)

Carlson and Knight: Agrilus beetles 21

found that pruning only one half of the parent root of aspen suckers causes lit-
tle mortality, even after two years. Consequently, horni-induced sucker mor-
tality must be considered to result mostly from girdling of the stem.

The extent to which horni larvae feed in the cambial region after returning
to the stem may be related to sucker vigor (Nord, 1962). Stager (1923) found
that Agrilus aurichalceus Redtenbacher (= rubicola communis Obenberger)
boring in bramble (Rubus spp.) stems increases or decreases the distance be-
tween spirals such that the stem does not die until the period of pupal maturity.
Apparently, the death of the stem is a crucial factor, because Stager says the
pupae require a dry chamber. This seems to be true for species in the anxius
group, as well. One might, therefore, expect variations in the position of
horni pupal chambers to be at least partially related to moisture conditions in
the stem at the time of its construction. Their pupal chambers are located an
average of 4.8 inches above ground, but the position varies from a height of
2 inches to 14 inches. (Nord et al., 1965).

POST-FEEDING LARVAL PERIOD

Formation of the Pupal Chamber. Brooks (1914) made a detailed analysis
of pupal chamber formation by A. vittaticollis. The vittaticollis larvae ex-
tend a burrow of the usual dimensions through to the bark at the upper end of
the gallery. They then recede about 1/4 inch and begin eating wood from the
sides of the gallery. As this proceeds, their feeding is directed backward
along the ventral side of the body, thereby doubling the gallery diameter. The
two ends of the body are, as a consequence, appressed closely together, and
the head and anus soon lie side by side at the bottom of the pupal chamber.
The chamber is then complete, but the larvae continue to move forward until
the head and anus are at its upper end.

The larvae of all Agrilus spp. probably construct the pupal chamber sim-
ilarly. However, when the layers of birch bark are gradually peeled off over
anxius pupal chambers, a small loop in the gallery is found at the upper end of
the chamber (exit point for imago). The loop apparently marks the point at
which the larva turned in the bark and re-entered the xylem. It appears to be
peculiar to anxius, and may somehow relate to the peculiarities of birch bark.
The loop is not evident at the exit points constructed by g. liragus larvae, and
horni and pensus, at least, must double back in the manner Brooks (1914) has
observed for vittaticollis.

Shortening of the larva may be initiated earlier in most of the other spec-
ies of the anxius group than it is in vittaticollis. Whereas vittaticollis initially
settles in the pupal chamber with the body folded in half, the larvae of other
species assume a position in which the body is folded between the second and
third abdominal segments. While vittaticollis has both head and anus at the
upper end of the chamber, the other species have the folded portion of the body
at the upper end, the anus at the lower end, and the head pointed downward ad-
jacent to the abdomen (in standing trees). This would indicate that instead of
stopping where vittaticollis does, with head and anus at the upper end of the
gallery, they move around farther. An alternative possibility is that after
boring to the outer layers of bark, they merely turn back and feed along the
ventral side of the body until the gallery width is doubled, then gradually re-
cede or shrink to the overwintering position. This seems unlikely, however,
as the exit point is plugged with frass. Unless these Agrilus have acquired
the ability to pack frass with the mouth parts and head, it must be assumed
that the anus, at some time, is at the upper end of the pupal chamber. Brooks

22 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

(1914) implies that it is the fourth instar vittaticollis larva which constructs
the pupal chamber, which may explain the difference in the folding position for
the abdomen in this species. The molt to the fifth instar occurs before forma-
tion of the pupal chamber in pensus, anxius, horni, and g. liragus, and the
fifth instar larvae are noticeably shorter and more stout than fourth instar lar-
vae.

Diapause. Barter (1947) has shown that it is necessary for anxius larvae
to reach the fifth instar before overwintering in order for pupation to occur the
following spring. He says that subjection of fifth instar larva in the pupal cham-
ber to subfreezing temperatures is necessary for the continuation of develop-
ment.

In early October 1966, some sections of a small birch tree infested with
anxius were collected near Manistee, Michigan, and stored indoors with the
intention of removing the larvae from them at leisure. In November part of
them were debarked, and none of the larvae had shrunk to the prepupal form.
In early December, however, adults emerged from these logs, and a few more
issued in late December. Some larvae, on the other hand, did not enter the
prepupal stage, suggesting that only some of them had been exposed to low
temperatures in the proper stage and for a sufficient duration before they were
brought indoors. This is in agreement with recent theories which, according
to Wigglesworth (1964), suggest that diapause involves some phase of develop-
ment that can proceed only at low temperatures.

While the transition from the overwintering stage to the prepupal stage may
be possible only after some exposure to low temperatures, the transition it-
self appears to occur only when temperatures are above Rie and probably
considerably higher. In Michigan the transition of anxius, g. liragus, and hor-
ni larvae to the prepupal stage occurs during the spring. In Benzie Co. the”
transition of liragus larvae, in some cases, occurs before late April, but usu-
ally is later. In Iron Co. some horni larvae do not go into the prepupal stage
until early June.

PUPAL TRANSITION. In Iron Co. the transition from the prepupa to the
pupa usually occurs in May and early June for anxius, g. liragus, and horni.
There is no delay in development when the pupal stage is reached, and soon
after the all-white pupa forms it begins to take on some of the adult coloration.
The eyes are the first parts to darken, and the mandibles darken next (Barter,
1957). The elytra remain white until most of the other parts of the body have
darkened, and do not come to rest in the adult position until rather late in pu-
pal development. Very likely, this is the reason that the wings are deformed
more frequently than other parts in rearing these beetles. When drying of the
wood around the pupal chamber is too rapid, the elytra may not take shape
properly, after uncurling from the sides of the abdomen. This usually results
in a poor fit of the elytra along the suture, causing a gap which exposes the
membraneous wings. Sometimes the deformities are more extreme.

ADULT EMERGENCE

Study Procedure. Eclosion of the adults from the pupal chamber was fol-
lowed in several ways. Unfortunately, no good field emergence data were ob-
tained for g. liragus, and no attempts were made to get records of field emer-
gence for pens

In 1964 g. liragus emergence began about May 30, which was considerably
earlier than was expected. Consequently, there was not sufficient time to lo-
cate enough pupal chambers for checking field emergence adequately. Some
data were obtained by placing small triangular screen cages (fig. 13) over the

Carlson and Knight: Agrilus beetles 23

exit points that were exposed by cutting away the outer bark layers. The same
method was employed in 1965, but because emergence was considerably delayed
the results were extremely poor. Dessication of the exposed bark layers and
wood prior to the beginning of the emergence period caused high mortality of
the pupae and young adults.

The same type of cage worked well for analyzing field emergence of anxius.
To expose the exit points for anxius adults, only a few outer layers of birch
bark had to be removed. Consequently, the rate of dessication of inner bark
and wood probably differed insignificantly from that of infested trees not trea-
ted in this way. A. anxius emerged normally from caged trees in 1964 and
1965. In 1964 the cages were checked intermittently, only often enough to
roughly establish the beginning, end and peak period of emergence. In 1965
they were checked daily, except on days when emergence was unlikely.

In 1964 the emergence of horni adults was followed in the field by examin-
ing some infested suckers marked with flagging. Although the plot was checked
only every other day, this method proved to be rather inefficient. Hence, dif-
ferent methods were used in 1965. The basal portions (sticks) of 123 infested
P. tremuloides stems were placed in a clearing at the University of Michigan
Forestry Camp near Beechwood, where they received full sunlight during the
middle of the day. The ends of the sticks were waxed, and the pupal chambers
were then located by comparing the sticks with their X-ray! negatives. Tubes
of plastic screening were positioned over the segments of the sticks harboring
the pupal chambers, and the ends of the tubes were constricted by stapling or
taping. The sticks were held in an upright position near the ground by cords
strung between stakes (fig. 14). They were checked daily.

It is probably safe to say that in most parts of their ranges, peak emer-
gence for anxius, g. liragus, and horni occurs during June, but even within
a given region there can be significant year-to-year variations in the emer-
gence pattern. The variations in emergence periods relate to weather varia-
tions. Davis and Raghuvir (1964) compared deviations from long term mean
air temperatures with emergence data for A. aurichalceus. The temperature
data used were those for the month of March, April, and May. Walton (1951)

Fig. 13. Small emergence cages stapled over exit
points for Agrilus anxius adults.

1
The X-ray machine and techniques used are discussed by Knight and Albertin (1966).

24 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Fig. 14. Plot at Camp Filibert Roth near Beechwood,
Michigan, where the emergence of Agrilus horni was
studied under semi-natural conditions by stringing the
infested sticks between stakes to hold them vertically.

employed the same technique for studying the emergence of A. ruficollis Fabri-
cius for successive years. Emergence appears to be sufficiently well corre-
lated with monthly temperature data to make the technique useful.

Results and Discussion. Table 1 shows the deviations from long term
monthly means for Stambaugh, Michigan for March, April and May in 1964
and 1965 (U. S. Weather Bureau, 1964).. The average deviation for 1964 was
+0.97 degrees, while that for 1965 was -1.23 degrees. Emergence data for
these two years reflect the differences.

In 1964 anxius emergence began in the Stambaugh vicinity about June 4 but
not until June 14 in 1965. The difference between peak emergence times was
even greater, being about June 7 or 9 in 1964, and between June 20 and 23 in
1965. In 1964 horni emergence began about June 7, but was delayed until a-
bout June 20 in 1965. Peak emergence was between June 9 and June 15 in
1964, while it occurred between June 21 and June 28 in 1965 (fig. 15).

The 1964 emergence periods of horni and anxius did not extend past the
end of June, but in 1965 both species continued emerging into July. Surpris-
ingly, one anxius adult was found alive in an emergence cage on August 4,
but with this unusual exception, emergence of this species ended by July 7.

A. horni emergence ended on July 15, making the duration of the 1965 emer-
gence period about equal to that for 1964.

Table 1. Average deviation from long-term mean
air temperature (F) for Stambaugh, Michigan.

eich 1964 co 1965
March “eae ~O. 6
April +13 “2. b

May ab ih +4.0

Mean +0.97 -1.23

25

Carlson and Knight: Agrilus beetles

‘arlojoq skep Z IO [ uUeY} 91OW poSi1oWIN pey }I JI peep
useq sAvY PTNOM pue F 4SsNsny UO oSvod sOUSS.IOWIS US UI BAITS PUNO} svmM }[Npe sNTxue ouGO (Zz) ‘UMOUYyUN ore soyep ooues
-IOULS OTJIoeds puv peYxdeYO oq JOU P[NOd j4oTd 9} [BAIOJUL STU TOF ‘POST ‘ouNL JO YI 6Z 94} YSno1Y} YIOT OY} Wor periMmd90
SooUdSIOWO TULOY SATA (T) “SUIMOT[OF 94} OJON ‘“UBSTIYOT|! ‘*OD UoA ul ‘dds sn{lisV Jo soyep soueS10UIN YJNpY “CT ‘BIW
i ec Pe ee 0

Gg6l ‘SnSeaT] SnyejnueIs SniIwy OI

©
re
Aed

ied Suisi9sw gy seljeog jo 19quINN

G96 ‘SnIxue SnTIIsV

0 bo

S

Ui) &
oD)

POEL ‘TULOY SNTIIsy OL

Pete wor tee Poe eS Pee ke 6 Ce eS ee ETS ORS PS eee Tee eh 9 eg

ATOL ANAL

26 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Barter (1949) found that the mean emergence dates for g. liragus preced-
ed those for anxius by an average of 12.7 days. Although data obtained for
g. liragus inour study were rather poor, it is evident that the differences in
time of emergence are not as great in Michigan. A. g. liragus began emerg-
ing about 7 or 8 days before anxius in 1964, but the peak emergence periods
were not as far apart.

Anderson (1944) reared both anxius and g. liragus from caged aspen and
birch logs in Minnesota. He considered the possibility that he was working
with two ‘‘host races, ’’ so one might assume that he would have noticed any
significant difference in emergence times. In combining the emergence data
for both anxius and g. liragus, Anderson found that emergence began on May
29 and ended on July 2, the peak occurring on June 10.

A difference in peak emergence times for g. liragus and anxius would ap-
pear to have no more real significance than any other differences between these
species, and may be subject to the same kinds of geographic variation. These
beetles are, in any case, actively mating and laying eggs during time spans
which are nearly coextensive.

Population Regulating Factors

The factors which are usually considered important in the regulation of
insect populations are parasites, predators, and weather conditions. There
have been many semantic arguments devoted to the question of whether or not
factors which cannot take an increasing toll on an organism as its density in-
creases can be as important as factors that can increase their toll with density.
Because factors like weather conditions have no way of responding to density,
it is usually assumed that they fall in the former category. More often than
not, though, weather conditions are clearly a part of the climatic scheme to
which a species becomes adapted. There may be some variations of weather
which occur so infrequently that they are not likely to bring about genetic adapt-
ation in the species affected. Such catastrophic factors are here considered
to be of secondary importance as population regulating factors; a species must,
in the long run, be part of a density dependent controlling system. The factors
likely to be of major importance in controlling the Agrilus being considered
here, are parasites, predators, and host resistance.

HOST CONDITION. Host resistance can be regarded as an indirect effect
of weather conditions, but it is probably more valid to consider it to bea
manifestation of the more or less balanced relationship between the host and |
its phytophagous parasite, and a manifestation of the physiological condition
of the host. The physiological condition of trees is most directly related to
their age and site quality. By loosely applying these two variables one could
explain most of the interactions which occur between Agrilus spp. and their
hosts but the balance of the interrelationship is also influenced by parasites
and predators, both phytophagous and entomophagous.

The Bronze Birch and Bronze Poplar Borers. Anderson (1944) says the
larvae of anxius or g. liragus require living ‘‘cambium’’ if they are to com-
plete their development successfully. Inner bark tissues which have died and
turned brown are unsuitable for larval development. Barter (1957) found that
some first instar anxius larvae may bore 13 to 18 inches without molting in
moribund phloem. Those that are able to molt bore about twice as far (6 inch-
es) as larvae feeding in phloem which has not turned brown. The evidence

Carlson and Knight: Agrilus beetles 27

seems to indicate that these larvae require nutrients associated with living
phloem and that these essential nutrients decompose when the phloem cells
die. |

| Swaine (1918) and Balch and Prebble (1940) reported finding anxius attack-
ing and killing many healthy, mature birches in Canadian forests. Nash et al.
(1951) reported that during times of high bronze birch borer population levels,
healthy trees are attacked and killed. Different opinions were held by Spauld-
ing and MacAloney (1931) and Hall (1933), who believed that the borer attacks
birches because they are ‘‘dying.’’Barter (1957) suggests that the borer may
appear to be more aggressive under outbreak conditions, but that it otherwise
plays only a secondary role. He and Anderson (1944) present evidence show-
ing that the beetles are usually unable to successfully attack healthy trees un-
less they are exposed to repeated unsuccessful borer attacks in successive
years. Of course, after such repeated attacks, the trees could not be con-
sidered healthy, and persistent attacks of this sort could only occur during
extended outbreaks. Perhaps unsuccessful attacks indicate that the number of
susceptible trees is limited, or that ‘‘spill-over’’ occurs from highly attractive
trees onto less attractive ones adjacent.

It is apparently only when many trees are simultaneously predisposed to
infestation that the bronze birch borer becomes epidemic. A. granulatus
liragus apparently does not often reach the outbreak state, but Barter (1965)
and Barter and Cameron (1955) found that following repeated defoliation by
the forest tent caterpillar, Malacosoma disstria Hber., aspens are predisposed
to g. liragus infestation. Under such conditions g. liragus can cause deterior-
ation and death of hosts it attacks either on branches or the main stem. Arru
(1962) discusses the factors which predispose newly transplanted poplars in
northern Italy to attacks of Agrilus suvorovi Obenberger (=viridis populnea
Schaeffer). Problems with this insect arose only after poplars were planted
on soils which were not as well suited to them as the deep friable soils to
which poplar cultures were formerly confined. Arru believes that the factors
which make the transplants susceptible to infestation by the beetle are those
preventing resumption of growth following planting. The most important of
them is moisture stress, which is likely to occur when poplars are transplanted
on sandy or pebbly soils, on embankments, or in compact, poorly aerated soil.

Anderson (1944) treated aspens and birches in a number of ways to deter-
mine the severity of injury, if any, necessary to predispose them to the attacks
of g. liragus and anxius'. He made some interesting observations on brood
development in 6 aspens and 6 birches which were girdled in September, 1940.
All of these trees, except one of the aspens, remained alive throughout the
summer of 1941, when they became infested with g. liragus and anxius. Brood
density below the girdles in the aspens averaged 1.2 per sq ft (52 sq ft exam-
ined), while above the girdles there were only 0.1 larvae per sq ft (88 sq ft
examined). Similar results were obtained for the birches. Below the girdles
all the living borers were either pupae or adults, but only larvae were found
above the girdles during the spring of 1942.

All of the trees produced a small annual ring above the girdle during the
summer that they were attacked by Agrilus. No growth rings were produced
below the girdles in any of these trees, however.

When Anderson (1944) debarked the girdled trees he found that the bark

Both taxa were still lumped under the name anxius at that time.

28 Contrib. Amer. Ent. Inst., vol. 4, no.3, 1969

moisture content below the girdles was nearly 2 times that above. He assumed
this to be a reflection of conditions that had existed during the previous summer.
This assumption is incorrect because girdling trees by cutting through the
phloem causes starvation and death of the roots. After the roots of a girdled
tree die, water absorption ceases and the rest of the tree probably dies slow-
ly from dessication (Kramer and Kozlowski, 1960). Because the phloem above
the girdle continues to receive small amounts of carbohydrates, while that be-
low gets none, the former should maintain a higher moisture content. It is
the presence of solutes (sugars, etc.) in the phloem which allows its turgidity
to be maintained. The diffusion pressure deficit of phloem sap offsets the
effects of negative pressure or tension in the xylem. Therefore, the high
bark moisture contents Anderson found below the girdles in the spring of 1942
were possible only because the bark had absorbed moisture during the winter.
Hence, he has used a false premise in asserting that the larval mortality in
trees with a proliferating cambium is not related to bark moisture content.

Heering (1956) found a relationship between bark moisture content and the
susceptibility of beeches (Fagus sylvatica) to attack by Agrilus viridis.
However, he claims that in some cases callose formation becomes the impor-
tant host resistance factor, especially in trees so extensively dessicated that
defensive sap flow is impossible. Because callose is a polysaccharide con-
sisting of linked glucose units (Kessler, 1958), its formation involves a proc-
ess much like normal growth. The growth of trees is probably controlled
more by water availability than by any other environmental factor (Kozlowski,
1964), and deficits in water availability can cause physiological conditions
which override growth regulatory processes (Zahner and Donnelly, 1967). It
is, therefore, difficult tosee how severely dessicated trees could form callose
fast enough to pinch off continuously feeding larvae. The most logical explana-
tion is that when Heering thought larval mortality was attributable to callose
formation, the larvae had, infact, been overwhelmed by phloem exudates, and
were afterward being overgrown by callose. Even his own observations sup-
port this contention. He found that in very moist, vigorous bark, the virdis
larvae construct vent holes (‘‘Rindensaft-Abflusskanalen’’) which prevent sap
from accumulating in the gallery by allowing it to drain off to the outside. The
very fact that the larvae have time enough to cease forward movement and
push the anterior part of the body through the bark, suggests that wound cal-
lose formation could not be rapid enough to kill them.

Heering (1956) found that the moisture content of bark infested with first
instar larvae is almost as high as that of sound bark. This could indicate
that bark which appears to be sound may really have low vigor, or instead,
that his technique is too crude for measuring the essential differences between
susceptible bark and bark resistant to infestation. He removed his bark sam-
ples from the trees, placed them in glasers’ putty, and analyzed their mois-
ture content in the laboratory. His index of tree resistance was bark mois-
ture content expressed as a percentage of its total dry weight. This measure
of host resistance is not precise because the active phloem constitutes a very
small fraction of the total bark mass. Critical differences in its turgidity
would be masked by variations in other bark factors, such as the amount of
accumulated cork or innactive phloem. Perhaps a refinement of the technique,
such as that used by Bier (1964) would give better results.

However, phloem pressure would probably be a better measure of active
phloem borer resistance for the majority of woody angiosperms. Some of the
angiosperms have gum ducts and may have resistance mechanisms analogous

Carlson and Knight: Agrilus beetles 29

to those of conifers. For some conifers a close relationship between oleo-
resin exudation pressure and tree resistance to scolytid attack has been demon-
strated (see for example, Vité, 1961). While the nature of conifer oleoresin
ducts and their products makes pressure measurements relatively easy, host
resistance could not be studied as readily for species whose chief resistance
mechanism is the exudation of phloem sap. Piercing the phloem causes an
injury which interferes with its normal activity. However, the observations
of Heering (1956) suggest that useful pressure measurements might be ob-
tained without tapping the active phloem itself. He says that even before the
heads of newly-hatched Agrilus viridis larvae have completely entered the
bark, they are often repelled by exuding sap which detaches the egg shell from
the bark. Without the shell to push against the larvae are helpless.

If, as it appears, phloem pressure is actually the most important host
resistance factor encountered by phloem feeding Agrilus it should follow that
the serpentine galleries, like those often produced by g. liragus and anxius
are an adaptation giving them some advantage for withstanding the copious
flow of sap. In starting a zig-zag or sinuate gallery pattern, a larva makes
an initial cross-grain segment, then turns abruptly and reverses its direction.
When it begins to bore past the lateral extent of the gallery segment to which
it is adjacent, the larva usually reverses its direction.

Heering (1956) believed that the sinuate galleries produced by Agrilus vir-
idis larvae are the consequence of a fixed pattern of behavior, although they do
not follow a sinuate course in very decadent hosts. Barter (1965), on the
other hand, found that g. liragus makes galleries with very close transverse
segments in relatively vigorous hosts, while in more decadent ones the gallery
becomes more sinuous resembling those of anxius. Like xylem departures,
which for anxius or g. liragus increase in number with increasing host vigor,
the zig-zag or sinuate gallery pattern may be elicited as a sort of avoidance
reaction.

There seems to be little reason for assuming that the effect of phloem sap
on the larvae could be anything other than simple suffocation. Presumably,
when the larvae make departures into the xylem, the plug of frass behind
them protects them from the copious influx of phloem sap. The advantages
which might be derived from a serpentine gallery pattern are more nebulous,
however.

Barter (1957) suggested that by staying beneath the previously formed parts
of the gallery, the larvae are protected from the downward, translocation of
sap. However, downward progressing galleries do not seem to prevail over
those progressing upward in standing trees. Barter (1957) knew that galler-
ies can proceed in either direction, and his hypothesis is, therefore, inade-
quate, even for explaining the facts as he knew them.

Moreover, there seems to be good reason for believing that movement of
sap in the phloem of even the most vigorous of trees which Agrilus can infest
is, at most, very slight. Trees susceptible to Agrilus infestation are not
healthy and, without exception, must be subject to high internal moisture
stress. According to the mass-flow theory, phloem transport occurs from
areas of higher osmotic sap concentration to regions of lower concentration.
During the growing season, the direction of transport would normally be from
the crown toward the roots. However, when trees are under severe moisture
stress, the rate of respiration in the crown may equal or exceed that of gross
photosynthesis. Under such conditions there may be no transport of carbohy-
drates in the phloem, and its turgidity must become significantly lower than

30 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

in trees not under moisture stress.

Furthermore, serpentine or zig-zag galleries are produced even in logs,
if they are infested rather soon after being cut. This is strong evidence sug-
gesting that they are produced in response to local turgidity rather than to the
mass flow accompanying the translocation of carbohydrates. It seems fairly
apparent that even when moisture stress is too great for the translocation of
carbohydrates to occur, the phloem is still sufficiently turgid for sap to in-
fuse into sieve elements that have been severed. When Agrilus larvae cut
through the sieve tubes, part of the exuding sap is probably absorbed by the
inactive phloem and the xylem. Consequently, the pressure of sap in gallery
should be lower than that in the surrounding phloem. In spite of the fact that
successful infestation must invariably be attended by low turgidity in the
phloem, sap should pass slowly from it until it loses its turgidity entirely or
until its pressure comes into equilibrium with that in the gallery.

If the foregoing premises are valid, the function of the sinuate or zig-zag
gallery pattern may be unrelated to phloem transport or to minimizing the a-
mount of sap that enters the gallery. It seems possible that they minimize
only the detrimental effects of sap oozing from newly-severed sieve elements,
as is illustrated in figure 16. When a larva turns back adjacent to the prior-
formed gallery link, the phloem inside the turn is cut off between the links
forming a peninsula. These peninsulas should not exude as much sap as the
phloem outside the turn. The minimization of sap influx from the peninsula
might make it possible for the larvae to minimize its suffocating effects by
opening the spiracles on the side of the body adjacent to the peninsula. With
each successive link, then, the larvae could open the spiracles on alternate

Fig. 16. Sinuate gallery pattern similar to that of Agrilus anxius.
The vertical arrows indicate the hypothesized direction of sap infus-
ion into the gallery after severing of the phloem sieve elements (no
active transport is occurring). Because the body of the larva is a-
pressed very tightly to the walls of the gallery, alternately opening
the spiracles on the side of the body exposed to the smallest amount
of sap infusion, might allow continual respiration.

Carison and Knight: Agrilus beetles 31

sides of the body.

The differences in the pattern of the g. liragus and anxius galleries may
be partially related to differences in the phloem structure of birches and pop-
lars. It is not sound evolutionary thinking to assert that the gallery differences
are attributable either to genetic differences in the insect species or purely to
differences in the hosts. In other words, it is pointless to try to separate the
combined phenotypic effects of environment and genotype.

A. anxius or g. liragus larvae in relatively vigorous hosts construct longer
galleries than those living in weaker hosts. Barter (1957) found that anxius
which matured in one season in rapidly dying birches, made galleries of only
15 to 20 inches, while those in more vigorous trees matured in two years, af-
ter boring 20 to 52 inches. Anderson (1944) gives similar data for g. liragus
development. The reason for the vigor-related variations in gallery length
are apparently unknown, but two possibilities are evident: (1) larvae feeding
in relatively vigorous tissues perhaps must expend more energy in attaining
the nutrients necessary for complete development, or (2) molting of late lar-
val stadia may occur only under certain conditions, such as those relating to
host physiology, time of year, or both. There are no published data from
which it is possible to ascertain if the gallery segments for all larval stadia
are longer in more vigorous hosts or if the increase in gallery length is built
up during a particular portion of larval development.

The Alder-Birch Borer. Very little is known about the relationship be-
tween Agrilus pensus and its hosts, Alnus spp. and Betula nigra. Because
it is often difficult to distinguish pensus-infested speckled alders (Alnus rug-
osa) from uninfested ones, it seem likely that pensus plays a more primary
role in causing the death of these shrubs than do g. liragus and anxius in in-
festing their respective hosts.

It is very difficult to trace out complete pensus galleries because they
meander extensively through the xylem and a number of galleries may inter-
sect. Consequently, no attempts were made to accumulate data on total gal-
lery length. It seems likely, though, that the average length of pensus gal-
leries will approach the maximum length for anxius or g. liragus galleries.
Because dying speckled alders are probably not suitable hosts for pensus,
these insects are always contending with a level of host resistance which is
found only in the more vigorous of the hosts suitable for anxius or g. liragus.
Moreover, many speckled alders show signs of unsuccessful attacks, and suc-
cessful ones do not seem to occur very frequently. Much of the earlier por-
tion of larval life is spent in the xylem, and the mortality rate of young larvae
seems to be very high.

Blanchard (1889) records one location in Massachusetts where he found
pensus (under the name granulatus) in rather large numbers. Barter and
Brown (1949) collected a large series of pensus on Alnus crispa var. mollis
and suggest that it is particularly well-adapted to this alder, which

is confined to the Atlantic seaboard. However, A. c. var. mollis does not
appear to differ greatly inform or in habitat from A. rugosa. Furthermore,
in June, 1968 an area in which pensus adults were moderately abundant was
encountered at Mallard Creek in Iron Co., Michigan. The creek bottom was
flooded by beavers during the previous summer and possibly also in the sum-
mer of 1966. It, therefore, appears that, at least throughout most of its
range, pensus becomes numerous only infrequently, particularly because of
the host resistance controlling factor.

The Aspen Root Girdler. Agrilus horni has different means for coping

32 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

with host resistance than have anxius, liragus, or pensus. First of all, rather
than attacking dead or dying hosts, it limits its attacks to those poplar suckers
small enough to be overwhelmed, and those on which eggs are deposited appear
to be of normal vigor. The vigor of aspen suckers must be affected by the
root girdling of horni larvae. Zahner and Debyle (1965) found that pruning ei-
ther of the parental roots of Populus grandidentata suckers about 4 or 5 years
old usually causes a significant reduction in the growth of the suckers over a
span of two growing seasons following treatment. Usually, horni larvae com-
pletely masticate the xylem in one of the roots in which they bore. This occurs
at a point where the diameter of the root is about 1/8 to 3/16 of an inch in di-
ameter, and its chief function may be to elicit the turn of the larva back toward
_the stem of the host (see page 20). However, because water absorption occurs
mainly in a zone 1 to 10 cm from the apex of a root (Kozlowski, 1964), the xy-
lem girdling may induce moisture stress or increase it in the stem of the host
and in the part of the root proximal to the girdle. If the xylem girdle occurs
in a branch of a main (or parental) root, or if it sometimes does not occur at
all, the survival of larvae may depend upon their ability to effectively girdle
the phloem, thereby causing starvation in the distal part of the root or pre-
venting food from reaching the stem and proximal part of the root from the
parent tree (ortet). There are probably definite size limits on roots which
can be effectively girdled in either manner.

In 1960 Nord (1962) measured the diameters of 150 horni-killed suckers.
He found that 84 percent of these had diameters of 0.5 inches or less. He was,
however, unable to demonstrate that horni shows a preference for small suck-
ers because a similar percentage of the suckers in the stand he sampled had
diameters below 0.5 inches. Studying the same area in 1962, Fowler (1963)
found that 71 percent of the infested suckers had diameters below 0.5 inches,
while about 76 percent of the healthy ones were below that size. Using regres-
sion and variance analysis, he compared the relationship of sucker height to
diameter for infested suckers with that for uninfested ones, and found no dif-
ference. His illustrations show the two sets of height-diameter data to be
strikingly congruent.

Fowler (1963) lists the numbers of infested and uninfested suckers in each
of the age classes from 1 to 6 years, and his data are presented in table 2.

Table 2. Age-class frequency distribution of healthy
and Agrilus-killed Populus tremuloides suckers:

Number of Suckers

“ee Healthy horni-Killed
1 0 1
2 3 13
3 19 58
4 BOs 51
5) 93 293
6 20 4
Total 150 150

“Modified from Fowler (1963).

Carlson and Knight: Agrilus beetles 33

When the two distributions are made comparable by subtracting one year from
the age of the healthy suckers, it can be seen that the age-class frequencies
of infested and uninfested ones are very similar. It must be remembered that
suckers which died at age X were actually selected by females at age X-1.
Hence, the suckers available to horni females in this area were from 1 to 4
years old. The maximum height of suckers in the area would have been about
6.5 feet. This is near the maximum size of suckers on which horni eggs have
been found.

The data of Nord (1962) and Fowler (1963) seem to indicate that the horni
females were ovipositing randomly with respect to sucker size. It appears
that when the eggs were laid, few of the suckers were too large to elicit ovi-
position, or that too few were sufficiently above the optimum size for survival
of larvae to make it apparent in the statistics.

A small experiment was initiated in 1964 to examine the effects of host re-
sistance on horni larvae. Four groups of 15 Populus tremuloides suckers
were chosen with the intention of introducing first instar horni larvae at their
bases. Two groups were composed of suckers beginning their second year of
growth, and the other two groups of suckers were starting their fourth grow-
ing season. They will be referred to as 2-year-old and 4-year-old suckers.

An area was picked in which the likelihood of having the experiment con-
founded by natural infestation was low, although this approach, admittedly,
has undesirabilities. A random sample of the heights of 2 and 4-year old
suckers was taken. Heights were measured with a leveling rod and were re-
corded to within the nearest one-tenth foot of the point where the previous
season’s growth terminated.

The mean heights were calculated for each age class. Within each age
group, height classes of +1 standard deviation from the mean were set as the
sizes for suckers to be chosen for the experiment. The resultant height val-
ues for 2-year suckers were 1.3 and 3.2 feet, while those for 4-year-old
suckers were 3.7 and 7.4 feet. These treatments will be referred to as
2-year-, 2-year+, 4-year-, and 4-year+, respectively. On selecting the suck-
ers for the experiment a +0. 2 foot leeway from the treatment value was al-
lowed.

Eggs for rearing the horni larvae were procured from 10 cheesecloth cages
which were placed over individual aspen suckers (fig. 17 and 18). A pair of
beetles was introduced into each of the cages, and the eggs were collected
periodically. The eggs were removed in groups by cutting away a portion of
the bark on which they had been deposited. After the eggs were removed ad-
ditional soil was placed in the can which enclosed the base of the suckers so
that subsequent oviposition would occur on smooth, uninjured bark.

The eggs and bark to which they adhered were placed in small incubators,
which consisted of 1 by 3 inch white pine blocks with five shallow depressions
made with a 1/4 inch drill (fig. 19). The eggs were placed in the depressions
and a microscope slide was fastened over the tops of the blocks with masking
tape. To keep the incubators moist, dampened paper toweling stapled to wood
strips was held in contact with the ends of the wood blocks by rubber bands.
The pan in which the incubators were placed was covered to prevent dessica-
tion.

A camel hair brush was used for transferring larvae. They were placed
in small petri dishes lined with moistened blotting paper and transferred to
the test suckers as soon as possible. The larvae were placed in small cavi-
ties cut in the bark about one inch above the sucker bases. The cavities were

34

Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Fig. 17.

Unassembled cheesecloth cage for Agrilus horni oviposition.

Fig. 18. Assembled cheesecloth cage for Agri-
lus horni oviposition.

Carlson and Knight: Agrilus beetles 35

Fig. 19. Wooden incubators for Agrilus eggs. Standard
microscope slides (1 X 3 inches) are taped over the tops to
cover the small depressions holding the eggs.

prepared with a sharp scalpel by making two horizontal incisions about one

eighth inch apart through the bark to the surface of the xylem. The incisions
were cut obliquely downward so that when the strip of bark between them was
pried out, the cavity formed would keep the wriggling larvae from falling out.

After the larvae were placed in the cavities, they were covered over with
tape. Test suckers were subsequently revisited for checking on the survival
of the larvae. Those which had died before boring into the exposed phloem
region were replaced. All dead larvae were replaced until all suckers har-
bored larvae which had started into the phloem. It was necessary to replace
seventeen larvae before all suckers were infested, although it was apparent
when the suckers were examined in 1966 that some of the larvae probably died
after boring a very short distance. Introductions were begun on July 25 and
completed on July 31, 1964. A few days following completion of the introduc-
tions, the cavities in the bark were filled with grafting compound to prevent
fungal infection or dessication at these points.

Evaluations were made by digging up the suckers and their roots on May 4
and 5, 1966. The larval galleries were traced by cutting into the stems and
roots with a sharp knife. The results are tabulated in table 3. It can be seen
that only one larva matured in a 2 year- sucker, and none completed develop-
ment in the 4 year+ suckers. Six of the larvae which eventually formed pupal
chambers were in the 2 year+ (2) and 4 year- (4) suckers.

The data are, of course, rather meager, but it is likely that survival in
the 2 year- and 4 year+ treatment was poor primarily because of the unfavor-
able conditions for the larvae in the roots. In 6 of the 2 year- suckers the
larvae had to bore in roots that were, perhaps, too small to allow successful
development. Larvae are known to have successfully matured after spiraling
over a span of only about 6 inches in small roots, but this is probably not al-
ways possible. Survival of horni larvae was probably impossible in three of
the suckers having roots of adequate size, because the roots were diseased
and decomposing at various points.

One of the 2 year- suckers had a very large root, about 1 1/2 inches in
diameter. Such large roots probably are not suitable for horni larvae either,

36 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Table 3. Development of Agrilus horni larvae in
selected Populus tremuloides root suckers.

Number of Larvae

Ny a
A cee Boring at Least| Boring More Completing
Short Gallery | Than 10 Inches| Development

ayr- 14 8 1
2yrt+ 14 8 2
4yr- 11 5) a
4yr+ 11 4 0

All 50 25 ti

“Suckers were in their second (2yr) and fourth (4yr) growing sea-
sons when larvae were introduced and had heights (+0. 2 ft; see p.
33) of 1.3 ft (2yr-), 3.2 ft (2yr+), 3.7 ft (4yr-), and 7.4 ft (4yr+).

probably because they cannot be girdled readily. Large root size was undoubt-
edly an important factor contributing to mortality in 4 year+ suckers too, but
some of the larvae in these suckers seem to have perished even before reach-
ing the roots.

All the information so far assembled indicates that horni can kill healthy
aspen suckers. However, there is an upper limit to the size of hosts it can
subdue. It also seems likely that, if other factors are equivalent, the suckers
on areas of good site quality are able to resist infestation better than those on
poor sites. The areas in which horni is likely to establish persistent popula-
tions are ones where suckers spring up in fields at the edges of stands of lar-
ger aspens or in areas where the aspen clones do not expand rapidly to fill the
intervening spaces. The site quality in such locations is usually low even in |
the former case, but is invariably poor in the latter. A. horni is usually not
found in areas of high site quality, and never becomes abundant on such sites.
Nevertheless, one should not assume that their scarcity in areas of high site
quality is directly related to the poor survival of larvae in these areas. The
horni adults seem to orient to open areas, and good sites usually produce
dense, closed stands. Moreover, the adults may have evolved this affinity for
open areas at least partly through the effects of genetic selection channelled
via differential survival of the larvae.

Poor aspen sites which are left rather denuded after fire, like the main
study area of Nord (1962) and Fowler (1963) are likely to be good horni habitats
after the aspens have suckered. The numbers of horni in such places should
decrease rapidly over a period of 6 to 8 years, however, Infact, in 1964,
approximately four years following the year of peak abundance, there were on-
ly a few horni-infested suckers to be found on the plot where Nord and Fowler
had worked. The largest infested suckers listed in their data had diameters of
1.0 inches (6 inches above base). A. horni eggs were not found on suckers
larger than 0.9 inches at the base. It appears as though host selection may be
the factor limiting the sizes of suckers which are actually infested. Again, the
crucial survival factor would appear to be root size, and this certainly corre-
lates rather well with stem size. Although host resistance may not directly
limit the upper size of suckers in which horni larvae can live (i.e., by simply
causing the mortality of those above the upper limit) one must presume that
adult host selection behavior is evolutionarily directed by the ultimate results

Carlson and Knight: Agrilus beetles Je

OL it,

PREDATORS. Woodpeckers are the only predators known to exert any pop-
ulation limiting effect on anxius and g. liragus. Their predation ong. liragus
is confined to standing trees (Barter, 1965), and the same is probably true for
woodpecker predation on anxius. Knight (1958) found that the feeding of wood-
peckers on bark beetles has the greatest effect when the beetle infestations are
heavy and infested trees are concentrated. Similarly, Barter (1957) found that
their predation an anxius appears to be more prevalent in heavily infested
trees. One might then expect that the effectiveness of this predation would be
rather variable depending upon the size and concentration of populations, al-
though other factors are probably also important. Nash et al. (1951) found
that the toll of woodpeckers on anxius varied from very slight in some areas
to as high as 30 percent in others. Barter (1957, 1965), however, found that
woodpeckers consumed as much as 50 percent of anxius brood, and removed
up to 40 percent of the g. liragus brood from their pupal cells. Anderson
(1944) says that about 10 percent of anxius and g. liragus prepupae were con-
sumed by woodpeckers.

Woodpecker predation is confined to the spring, apparently occurring only
after the larvae begin the transition to the prepupal stage. Barter (1957) be-
lieved that woodpeckers fed on anxius at that time only, but later (1965) he
found that the pupae of g. liragus were consumed and presumed that adults
were also eaten.

Woodpeckers are also an important population limiting factor for pensus.

It appears that the predation on pensus occurs during the winter. Large num-
bers of pensus brood are never found in a single Alnus rugosa stem, but the
extent of concentration of infested alders may affect woodpecker activity. It
seems that once a woodpecker finds the first pensus pupal chamber in a stem
he is likely to find them all, although there usually are not more than 7 or 8
in a single stem. | :

We have seen spiders and ants attacking the adults of g. liragus and anxius,
but they probably cause little mortality. These beetles usually apress the legs
to the thorax and fall free of the host in time to avoid being caught. In one in-
stance, however, an ant had grasped its falling prey and both predator and
prey fell to the ground where they couldn’t be found. The quickness with which
these beetles take flight or drop to the ground, especially on hot days, is ap-
parently reflected in Gory’s name anxius (nomen nudum of Dejean), and it
seems rather unique that torpidus Leconte has been placed as a synonym of the
former.

Nord et al. (1965) observed ants mouthing the eggs of horni on caged aspen
suckers. Since ants frequently pass up and down the stems of suckers, often
in visits to sources of aphid honey dew, they may encounter the horni eggs
which are placed above ground level. Because many eggs survive to hatch,
they do not appear to be particularly attractive to ants, however.

PARASITES. Nash et al. (1951) listed eight parasite species which were
reared in cages containing logs infested with anxius. It is certain that some of
the Ichneumonidae they list as anxius parasites were instead reared from other
insects’. Glypta sp. and Pimplopterus sp. are parasites of Lepidoptera, while

1
Identifications made for Nash et al. were undoubtedly reliable as they say these were
made by Townes and Muesebeck, both prominent specialists.

38 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Olesicampe sp. is a sawfly parasite. None of these could have been reared
from anxius.

The parasite Barter (1957) reared from birches was identified for him as
Ephialtes sp., but as more recently been identified as Dolichomitus messor
perlongus s Cresson'. The Ephialtes species reared from Populus by | Populus by Barter
(1965) is probably the same species. The parasite identified for Nash et al.
(1951) as Ichneumon sp. is probably Dolichomitus sp., perhaps D. messor
perlongus. Nomenclatural corrections in the usage of Ephialtes Schrank and
Ichneumon L. have reinstated their application to groups parasitic on Lepi-
doptera?.

Townes and Townes (1960) record D. messor perlongus as being reared
from Saperda calcarata Say, and S. concolor Leconte, and it also parasitizes
Oberea shaumii Leconte. Townes? expresses doubt that females of messor
perlongus could develop on Agrilus spp.

The records of several braconids reared by Nash et al. (1951) from anxius
are probably legitimate. The species are Atanycolus charus Riley, Spathius
simillimus Ashmead, and Doryctes atripes Provancher. They also reared
the chalcid, Phasgonophora sulcata Westwood.

Barter (1957) reared an additional braconid from anxius, Helconidea liga-
tor Say, but did not rear Doryctes. However, he records Doryctes sp. para-
sitizing g. liragus (Barter, 1965). Besides the Ephialtes sp. already discussed,
other parasites Barter reared from g. liragus were Atanycolus charus, Phas-
gonophora sulcata sulcata, and the eulophid, Tetrastichus sp., listed as probably be-
ing near rugglesi ieglesi Rowher.

Barter (1957) says that Atanycolus and Phasgonophora were the most im-
portant larval parasites of anxius, parasitizing 1 to 52 percent (average 14
percent) and 1 to 8 percent (average 4 percent) of the larvae, respectively.
Other parasites were recovered only occassionally. He recovered parasites
from only 9 percent of all larvae in his samples. Therefore it appears that
total parasitism was not usually as great as is indicated by the percentages
listed above, which were calculated for a sample of 600 larvae.

The only parasites which was consistently reared from g. liragus larvae
by Barter (1965) is Phasgonophora sulcata, which parasitized from 2 to 20
percent of the larvae. Other parasites were reared less frequently.

Parasitism on larvae of anxius and g. liragus may amount to a rather in-
significant portion of the total mortality factors which regulate their popula-_
tions. The same cannot be said for egg parasitism, however. Barter (1957,
1965) and Nash et al. (1951) found that parasitism was consistently high on
eggs of anxius and g. liragus, usually being near 50 percent for both species.
Apparently, two chalcidoids were about equally important, even though in in-

i
The specimens reared from Betula were lent to us by Mr. Barter for identification, but
those he reared from Populus has been accidentally destroyed.

2
The revision of the Ephialtinae by Townes and Townes (1960) had led to the change for

Ephialtes, and Ichneumon, sensu Townes (1944), included Dolichomitus, sensu Townes
(1960).

S
Personal communication, 1966; H. K. Townes, American Entomological Institute, Ann
Arbor, Michigan.

Carlson and Knight: Agrilus beetles 39

dividual cases the species were alternately dominant. One was identified as
Thysanus sp. (Thysanidae) and the other is an encyrtid near Coccidencyrtus.

While the egg parasites of anxius and g. liragus must be more important
in population regulation than the parasites of the larvae and pupae, the reverse
seems to hold for horni. Only four eggs in a sample of 147 (about 3 percent)
taken in 1965 were found to be parasitized. The parasite was identified as
Oobius sp.

Although they are exposed to parasites for a shorter time than the larvae
of g. liragus or anxius, horni larvae are parasitized much more heavily. A.
horni pupae are also parasitized rather frequently. Total parasitism of lar-
vae and pupae was about 42 percent in 847 rearings, with Tetrastichus nordi
Burks being reared most often. This eulophid is quite small and probably
lays its eggs on the larvae when they spiral just beneath the bark in the stem,
between the time of their emergence from the root and formation of the pupal
chamber. This was deduced from observing the behavior of a few T. nordi
females in the laboratory and from X-ray pictures of T. nordi larvae develop-
ing just beneath the bark on larvae that never formed a pupal chamber. Usually,
however, the horni larvae are not killed by Tetrastichus until after the pupal
chamber is formed.

Emergence of T. nordi adults corresponds with the time when horni larvae
are re-entering the sucker stems (August and early September). An average
of 49.14 17.1 adults emerged in a sample of 49 broods. The size of the adults
and their sex ratio is quite variable.

Three ichneumonids were found with about equal frequency in horni pupal
chambers. They parasitized a combined average of approximately 13 percent
of the horni collected. These parasites have longer ovipositors, enabling
them to place eggs on the larvae and pupae within their pupal chambers. Xylo-
phrurus agrili Viereck has been reared from horni larvae and pupae in both
spring and fall. Brooks (1914) says this species has two generations per year.
He found it to be an important parasite of A. vittaticollis, which bores in the
roots of various trees and shrubs of the family Roseaceae. In the fall it lays
its eggs on the vitaticollis larvae at about the time they are forming the pupal
chamber. The spring brood attacks pupae or larvae just prior to pupation.

A. vittaticollis may be the primary host for X. agrili, as Brooks found it to
destroy from 25 to 40 percent of the larvae and pupae of this beetle.

Xylophrurus bicolor maurus Townes emerged from horni pupal cells only
during the spring, and all of the collecting and rearing records which Townes
and Townes (1962) list for X. bicolor indicate that the adults fly only during
the spring. X. bicolor seems to have a diapause similar to that of many Agri-
lus spp., requiring an extended period of cold for development to be completed.
It is perhaps for this reason that efforts to rear the adults from suckers col-
lected during the initial stages of larval development are not very successful.

No rearing records were previously available for X. bicolor maurus. How-
ever, the western subspecies, X. bicolor bicolor has been reared from an in-
sect in Salix sp. and from Saperda moesta Leconte infesting Populus balsami-
fera ssp. trichocarpa (Townes and Townes, 1960).

The third ichneumonid, Mastrus smithii Packard, was incapable of emerg-
ing from horni pupal cells. The only adults obtained were removed from the
infested suckers where they died, often after boring only part of an exit hole.

potdrniined by B. D. Burks, U. S. Dept. Agr., A.R.S., Entomology Res. Div., Wash-
ington, D. C.

40 | Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

The mandibular teeth of these specimens are usually worn considerably. This
species is known to be a secondary parasite on the ichneumon fly Gambrus
extrematus Cresson and has been reared from the coccoons of two sawfly
species (Townes, 1951). It seems likely that it is poorly adapted for emerg-
ing from Agrilus pupal cells, and probably parasitizes ichneumonid parasites
of other insects with greater success, particularly those in the coccoons of
cecropia moths..

Besides Tetrastichus nordi, several other Chalcidoidea' were reared from
horni pupal chambers. They caused about 6 percent mortality of the horni
reared in this study, and this was attributed principally to Lampoterma sp? |
a pteromalid. This parasite emerges about mid-May, and therefore, oviposits
chiefly on prepupae or pupae of horni. From 3 to 10 adults develop in each
horni pupal chamber. This Lampoterma sp. appears to be limited to one
generation per year by an obligatory pupal diapause.

Eurytoma magdalidus Ashmead (Eurytomidae) was reared from several
horni pupal cells, but this species may be a secondary parasite. In two in-
stances it is known to have emerged from pupal chambers which harbored two
different types of parasite larvae.

Some of the larvae and pupae which died for unknown reasons (17.7 percent)
may have been killed by parasite stings. Consequently, the impact of parasi-
tism may be underestimated.

ADULT BEHAVIOR
Feeding and Maturation

Barter (1957) found that Agrilus anxius females, maintained in the labora-
tory, fed about 6 days before showing any tendency to lay eggs. Adults of
anxius and g. liragus are not found mating or ovipositing in the field until about
2 weeks following emergence. There is no reason to suspect that the same is
not true for horni and pensus.

Oghushi (1963) found that under laboratory conditions Agrilus auriventris
auriventris Saunders females formed ova 5 to 6 days after emergence, but full
grown ova were not present until 8 to 10 days after emergence. Oviposition be-
gan on or after the eleventh day following adult eclosion. Ogushi found no full-
grown ovaries in females which had not fed. Their development was very slow
in unfed beetles, many of which did not live long enough to reach the oviposi-
tion stage.

Similar developmental requirements are probably common to most, if not
all, Agrilus. Some unfed anxius females dissected quite soon after emergence
had partially developed ova. Without food, however, these females are prob-
ably unable to form fully-developed eggs, and usually die after a rather short
time.

A. anxius adults were maintained successfully during the winter on pieces

Ata nne by B. D. Burks, U. S. Dept. Agr., A.R.S., Entomology Res. Div., Wash-
ington, D. C.

Nord et al. (1965) list this species as Metastenus sp., but the application of Metastenus
has recently been changed (cf Burks, 1967).

Carlson and Knight: Agrilus beetles 41

of apple, but in nature their feeding is, no doubt, restricted to leaves. Hutch-
ings (1923a) says the bronze birch borer feeds more readily on the foliage of
poplars and willows than on birch. Barter (1957) found that anxius adults con-
sume the leaves of poplars, particularly those of Populus tremuloides, more
readily than those of Betula papyrifera. The findings of Nash et al. (1951)
differ slightly; in their laboratory studies anxius showed the greatest prefer-
ence for P. grandidentata foliage. They rate the species used in their tests

in the following order of decreasing preference: P. grandidentata, P. tremu-
loides, Salix spp., Betula lutea (= alleghaniensis), B. papyrifera, P. balsam-
ifera, and Alnus spp.

Very little is known about the feeding habits of anxius adults, but they were
occasionally observed feeding on B. papyrifera leaves by Barter (1957). Brit-
ton (1922) says bronze birch borer adults (those attacking birches) fly about
and feed to some extent on the foliage of willow, poplar, and probably birch.

We have not tried feeding horni adults anything besides P. tremuloides
leaves in the laboratory, and horni was found in the field on no plant other
than tremuloides. Aspen leaves undoubtedly comprise the major portion of the
adult horni diet, which is probably true for g. liragus also.

A. pensus, on the other hand, did not readily consume the foliage of as-
pens. Because they have been collected on the foliage of alders, but not on
other plants, one would expect them to show some specificity for alder leaves.
However, the only foliage which they seemed to eat readily in the laboratory
was that of Salix spp. Very little testing with alder was possible, however,
as the beetles were maintained in Ann Arbor, where alders are not readily
available. Usually, pensus adults did not survive for more than a few days,
no matter what kind of foliage they were provided. Adults of other species
kept in the same controlled environment chambers survived exceptionally well
(at 70 percent relative humidity; day temp 75°, night temp 60° F). The poor
Survival of pensus adults may have resulted because of a failure to provide
the proper temperature and humidity for pupal development.

Dispersal

Nothing has been written about the dispersal capacities of g. liragus or
anxius, and little is known about the dispersal of any Agrilus, including the
ones that are naturalized pests in the United States. Heering (1956) has ap-
parently made the only attempts at release and recapture studies with marked
Agrilus. However, as only 200 A. viridis adults were released he was unable
to recapture enought to give a reliable result.

Although Agrilus beetles may be capable of flying considerable distances,
they probably do so only rarely. In rearing and maintaining these beetles in
the laboratory we have seen nothing which would indicate the existence of any
sort of antagonistic mechanism by which ‘‘flight exercise’’ acts as a releaser
for feeding or reproductive activity as has been demonstrated for some aphids
(Kennedy and Booth, 1963) and scolytid beetles (Graham, 1959, 1961). If they
flew only far enough to encounter suitable hosts, they would usually be staying
quite near the places where they emerged.

Host Selection

Following the period of reproductive maturation the adults are attracted

42 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

to the hosts which have the particular attributes making them suitable for mat-
ing and oviposition. Many Agrilus and other Buprestidae seem to show a strong
tendency to select hosts which are well exposed to sunlight. Consequently,
those which live on trees are most active at forest edges or in park-like stands.

ANXIUS AND G. LIRAGUS . Generally, anxius females are said to oviposit on
the sunny sides of trees and on those most exposed to sunlight. However, Bar-
ter (1957) found that the photopositive tendencies of anxius can be overridden,
to some extent, by other stimuli, presumably olfactory ones. Barter says that
when the shady sides of trees are wounded, females oviposit there rather than
on the unshaded side. He concluded that the olfactory response is often strong-
er than the photopositive response.

Anderson (1944) came to similar conclusions. During the first season of
his two-year study he found no deviation from the commonly accepted idea that
trees exposed to full sunlight attract significantly greater numbers of anxius
and g. liragus than shaded ones. Using a 30 second counting interval Ander-
son found an average of 2.6 beetles per unshaded tree, while only 1.8 were
found on shaded trees. When the experiment was repeated the following year,
however, he found that the count for unshaded trees did not differ significantly
from that for shaded trees. During the first season, trees with the tops bro-
ken off by winter damage were abundant, but such damage did not occur during
the second year. Anderson thinks that greater numbers of weakened hosts
were available during the first season, allowing g. liragus to make a selection
for those more exposed to sunlight. His conclusion is that the physiological
condition of the host is more important than its degree of exposure to sunlight.

There may be sexual differences in the photopositive response, but little
has been done so far to account for the possibility. Males and virgin females
may be more photopositive than ovipositing females. Females ovipositing on
fallen trees are apt to work along the sides of stems on days when males orient
more towards the upper surface.

One is prone to confuse the effects of light and temperature. It seems quite
likely that these beetles will appear to be more photopositive on relatively cool
days (but still above 70° F) than on hot days. An observation made in 1964,
during studies of anxius behavior, illustrates this point. In order to concen-
trate the beetles on bolts cut from freshly-felled birches, the tops and other
unused parts of the birches were pulled well into the shade. This slash, never-
theless, attracted some of the beetles, and the bolts under observation, which
were in full sunlight, attracted only moderate numbers of them. That day the
temperature maximum was 85° F (U.S. Weather Bureau, 1964; Stambaugh,

15 July) and the sky was clear, so that temperatures may actually have been
too high for the insects to remain for extended periods in the direct sunlight.

In another of Anderson’s (1944) experiments, the attractiveness of 7 types
of host material was evaluated by tallying the numbers of adult g. liragus on
each aspen test tree during a 30 second counting interval. Observations were
made between 10 AM and 3 PM on sunny days when the shade temperature was
between 75° and 90° F. His results are presented in table 4, and it is quite
evident that injured trees are more likely to be attacked than uninjured ones.
Barter (1957) found this to be true for anxius also. From Anderson’s data
(table 4) one may also surmise that standing trees are more attractive to g.
liragus than similarly treated trees which are felled. Both the topped trees
and topped and girdled trees attracted significantly greater numbers of beetles
than trees which were felled. Perhaps this occurred because the standing trees

Carlson and Knight: Agrilus beetles 43

Table 4. Attractiveness of various types of Populus tremuloides
host material to Agrilus granulatus liragus.

Condition of Average Number of Beetles | Number of
Host Trees per Tree per Observatior “| Observations

1. Vigorous 0

2. Suppressed 0 21
3. Girdled 9 sl. 54
4. Topped | Lied 54
5. Topped and girdled 2.8 27
6. Felled and lopped 1.0

7. Felled, not lopped £3

“Modified from Anderson (1944). :

’The values are corrected so that each represents the number of bee-
tles observed per 17 ft of bark area, the average for all test trees.

“The only statistically significant differences are those between num-
bers not included within the span of a single vertical line.

were more fully exposed to sunlight than felled ones. In any case, these re-
sults could be misleading because standing trees rarely are as decisively in-
jured by natural factors as are trees which are artificially topped or topped and
girdled. The only natural injury which is really comparable occurs when trees
are blown over. Less decisive natural injuries occur to standing trees when
the tops are broken by winter ice damage or when the tops of already decrepit
trees break where the wood exposed by repeated insect attack and disease be-
comes dry and brittle. Aspens are frequently victims of this type of injury,
since the wood of these species is not particularly tough, and because they are
often severely attacked by wood borers like Saperda calcarata and fungal para-
sites like Hypoxylon pruinatum for a period of years before they finally
succumb.

One might also conclude from the data of table 4 that both suppressed and
vigorous trees are unattractive to g. liragus. In testing this more thoroughly,
Anderson used an extended observation period of 14 days, to detect occasional
beetles that might be attracted to the trees. This was facilitated by placing
sheets of fly paper on the trunks of 10 paired aspens, each pair composed of
a vigorous and a suppressed tree. During the 14 day observation period 18
beetles were caught by the fly paper on suppressed trees, but none were caught
on vigorous trees. It is apparent that g. liragus is able to detect trees which
are in poor condition, whether or not they have been wounded. The stimulus is
undoubtedly olfactory, and is presumed to arise from the decomposition of sub-
stances in the bark or phloem. The attractiveness of the host may, in turn,
depend upon the rate of decomposition, which is no doubt enhanced by injury.

HORNI. Theoretically, the kind of host selection exhibited by anxius and
g. liragus should allow adult feeding to be separated from the activities involved
in host selection. If this is true, then there might not be any highly developed
specificity shown by feeding adults. In the late 1800’s, buprestid collectors
clearly recognized that the identity of an Agrilus larval host was not necessarily

ag Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

indicated by the kind of foliage the adults are taken on. Perhaps this point is
over-emphasized, because the adults of some kinds of species, like horni,

may show more specificity in feeding, because feeding may be involved in the
host selection process. Agrilus horni does not fly directly to the stems of as-
pen suckers, but instead lands on the foliage. The adults of this species usu-
ally feed before moving onto the stem where mating and oviposition occur.
Whether or not they always feed before crawling from the foliage to the stem

is not known. Nord et al. (1965) found that after caged horni females oviposit
they move up to the foliage of suckers, resting and feeding for about 10 minutes
before decending to lay another egg. Under natural conditions the females may
usually ascend the stem following oviposition and fly to another sucker before
feeding.

A. horni can maneuver with agility in flight, and can hover before landing.
Their flight behavior indicates that they probably recognize aspen foliage be-
fore landing on it. If so, olfaction is clearly part of the recognition process.

If the odors which direct them to the host do not, at least in part, emanate
from the foliage, one would expect them to land frequently on the foliage of
plants other than Populus spp. This was observed only once, when a male
which had just mated flew from P. tremuloides sucker to a fern next to it,
staying briefly before he flew back to the same aspen sucker and began feeding.

To go no farther than suggesting that horni recognizes its host by some
characteristics of the foliage would be forming a hypothesis of very little
value. Obviously, other factors must strongly influence the host selection of
these beetles, as they do not live on just any species of Populus, or just any
P. tremuloides sucker. |

Previous Studies. Fowler (1963) studied the relationship of height, age,
and stand density of aspen suckers to infestation by horni. He counted the num-
ber of infested and uninfested suckers on systematically located millacre quad-
rats. A graph of Fowler’s data (fig. 20) appears to indicate that the unit-area
density of horni-infested suckers is rather directly related to the unit-area
density of all suckers. However, it is not necessarily true that unit-area
densities per se of the aspen suckers were the important variables affecting
the infestation levels throughout Fowler’s study area. It seems likely that some
of the lower stand density values can be attributed to parts of it where the ini-
tial establishment of suckers was poor because of severe competition with
other plants which dominated. Besides patches of sweet fern (Myrica asplen-
ifolia [=Comptonia peregrina]), leather leaf (Chamaedaphne calycultata), and blue-
berry (Vaccinium spp.), there were young red pines (Pinus resinosa) scattered
about. Quadrats with high sucker counts, on the other hand, may simply have
occurred in places where some of the suckers were not shaded by other plants.
In fact, high-count quadrats probably fell where suckers were clumped (fig. 21),
and sucker clumps probably had originated in spots that were left denuded af-
ter the 1957 fire. Fowler (1963) notes that when infested suckers were associ-
ated with clumps, they were usually at the outer edges of them. A majority
of the infested suckers in this may have occurred at the outer edges of clumps.
Therefore, the contagion evident in the distribution of infested suckers may
have resulted from strong contagion in the distribution of suckers suitable for
Oviposition.

_ much more detailed analysis of the data is presented by Carlson (1968).

Carlson and Knight: Agrilus beetles 45

Infested Suckers per Millacre Quadrat

aR ae a a ia ie Ce LE age oe ae? ae
Total Suckers per Millacre Quadrat

Fig. 20. Unit area density of horni-infested Populus trem-
uloides suckers as a function of stand density of all tremuloid-
es suckers in the area sampled by Fowler (1963).

New Studies

Methods. The possibility that some contagion in egg distribution could
occur independently of the distribution of suitable hosts has not been over-
looked. In the spring of 1964 an area with an evenaged stand of aspen suckers
was selected for studying this problem. It is an area near Beechwood, Michi-
gan, where furrowing for the planting of pines had caused prolific root sucker-
ing by scattered, mature Populus tremuloides, some of which were still stand-
ing, while others had been pushed over (fig. 22).

Because the problem involved determining the spatial distribution of in-
fested or egg-bearing suckers in relation to each other, there was little need
for using standard sampling methods for estimating density (i.e. stems/unit
area). Instead, it was decided that the problem could be best approached by
using distance measurements as indicators of spatial relationships. The two
variables measured were: (1) the distance from predetermined sampling points
to the nearest infested or egg-bearing sucker (DNS) and (2) distance between
each infested or egg-bearing sucker and the nearest neighboring infested or
egg-bearing one (DNN). Sampling points were located systematically. Five
plot centers were established, four at the corners of a 400 foot square and one
in its center and also approximately centered within the study area.

Around each of the plot centers, 8 sampling points were established ona
_ circle with a 50 foot radius. Sampling point number 1 was always located due
north of the plot center, and the others were located at subsequent azimuthal
intervals of 45 degrees. Sample distances were measured with a 100 foot tape.

46

Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Fig. 21. A spot in the study area of Nord (1962) and Fowler (1963), showing
the clumped distribution of Populus tremuloides suckers and abundance of com-
peting vegetation (from Nord, 1962). The aspen suckers occurred under similar
conditions throughout much of the area.

Fig. 22. Agrilus horni study area near Beechwood, Michigan. The stand of
Populus tremuloides suckers, which is relatively uniform in distribution,devel-

oped following furrowing for the planting of pines. Equipment used in the furr-
owing operation pushed over many of the mature aspens.

Carlson and Knight: Agrilus beetles 47

The Wilcoxon matched-pairs signed-ranks test (Siegel, 1956) was used for ana-
lyzing the data. Discussion of the results is deferred until later, where the
results of a number of analyses are discussed together.

While evenaged sucker stands are best for studying the distribution of eggs
by females, independent of sucker distribution, they are not well suited to
studying the sucker size selection of hornifemales. For this, unevenaged
areas are more appropriate. Unevenaged sucker areas occur in the ecotones
between aspen forests and open fields or in park-like stands where clones or
multiclone groups are apt to stand isolated, annually producing some suckers
at their edges. Such areas are also of particular interest because they har-
bor more perpetual infestations than do evenaged stands. Devising sampling
techniques for unevenaged areas was problematical, however. The most per-
plexing difficulty is that in these areas the zone in which the horni infestation
occurs represents a rather small proportion of the ecotone area. Hence, it
is difficult, if not impossible, to devise a random sampling scheme that is
efficient. Even a systematic approach is difficult to contrive because the habi-
tat of interest is usually distributed in a rather irregular manner. Therefore,
the usual kinds of systematic procedures are not applicable.

The compromise devised cannot be considered ideally objective, but that
does not seem to greatly impair its value. The sampling was carried out by
proceeding along the edges of the clumps or closed stands of larger aspens
and selecting the first 6 ft sucker encountered as a plot center. In turn, the,
Tit, 5 ft, 4 ft, 3 ft, 2 ft, and 1 ft suckers nearest the 6 ft one were selected.

The distance between the 6 ft sucker and each of the others (D,) was meas-
ured. Before the suckers were chopped free at their roots and examined for
eggs, the distance to the nearest neighboring sucker in the same height cate-
gory (DNN) was measured and recorded. Only the first set of suckers selected
was examined for horni eggs. Besides tallying the numbers of eggs and noting
their positions, we also recorded the basal diameters of the suckers (DAB).

Successive plots were always separated by at least 3 times the maximum
D,. Six foot suckers were chosen for the sampling centers because they were
near the upper size limit for suckers which would bear horni eggs. Conse-
quently, choosing a 6 ft sucker would insure that sampling would include areas
where suckers above the upper size limit occurred, while still keeping it with-
in areas that were likely to be horni habitat.

This sampling method was employed in one area near Kenton, Michigan
and in another near Phelps, Wisconsin. Analyses of the data chiefly involved
the use of chi-square comparisons and graphic procedures. Expected values
(E,) for the chi-square analyses were derived from expectancy factors (E,),
which are weighting factors based upon D, and DNN (collectively, D values)
measurements. The basis for the Seen ent ion of expected values is the as-
sumption that the abundance of suckers ina particular size class is inversely
related to the area of a circle which has a radius equal to the average D value
for the class. Hence, density is proportional to 1/ D7. In deriving the E, val-
ues it was anticipated that DNN measurements should account for any clumping
tendency among suckers of the same age. On the other hand, the D, measure-
ments should account for any stratification that existed, such as gradients in
the positions of suckers relative ,t0 those of other sizes. E, values are com-
puted simply by dividing the 1/ D? value for each size class be the total for all

1
Each height class had a+0.5 ft interval.

48 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

classes (giving each E, relativity to all others). The expected numbers of egg-
bearing suckers are derived by multiplying the total observed number of egg-
bearing suckers by each E, value.

One set of chi-square comparisons was made for height classes, and in
another set the numbers of egg-bearing suckers were apportioned into DAB
classes (table 6). Diameter class (DAB) values for the latter comparisons
were computed from regressions of DAB over sucker height by substituting the
values for the height class limits into the equation and solving for DAB.

To compare height and DAB as indicators of the suitability of suckers for
horni oviposition, ratios of observed to expected values are used. This re-
quired the computation of additional expected values for a set of DAB classes
with uniform limits. It was necessary to derive expected values for these
uniform DAB classes indirectly, because there was no direct (significant) sta-
tistical relationship between DAB and D values.

In order that all ratios of observed to expected values have positive signs,
it is necessary to adjust all expected values (E,) upward so that none is ex-
ceeded by the observed (O,) value for the same class. What was done, in
effect, was to multiply all E, values by the greatest O,/E, ratio occurring
among all sucker classes (yielding E,.). The subsequently derived O, /B,:
ratios can be used for making between-area data comparisons as well as for
the height-DAB comparison mentioned above. Factors with similar uses are
derived by dividing the total number of eggs (S) occurring on the suckers of each
size class by the corresponding E,, values.

Discussion. There were no significant differences between DNS (dis-
tance to nearest sucker) and DNN (distance to nearest neighbor) measurements
for either the infested or the egg-bearing suckers at the Beechwood, Michigan
evenaged sucker stand. Therefore, there is no marked tendency toward con-
tagion in the distribution of either infested or egg-bearing suckers. Two ob-
vious explanations are possible. (1) A. horni females may make a post-ovi-
position flight, before they re-alight and lay another egg. This sort of behavior
should lead to a rather random distribution of eggs. (2) On the other hand,
suckers which were suitable for oviposition and subsequent infestation may
have been rather randomly distributed throughout the area.

The latter explanation seems to be the most logical. The few direct field
observations made on horni females would indicate that they really do not fly
very far before landing on another sucker. Nor is there any indication that
a post-ovipoisition flight is an unvarying occurrence in a fixed behaviorial pat-
tern. Nord et al. (1965) found that after ovipositing, caged horni females im-
mediately crawled up the sucker stem and out onto a leaf. After spending about
10 minutes feeding and ‘‘resting,’’ they might descend to lay another egg, and
SO on. :

Under natural conditions horni females may usually fly to another sucker
after each egg is laid. It may be uncommon for them to feed on a sucker just
after ovipositing on it, particularly if feeding is an integral part of the host
selection process. In the Beechwood area only a few suckers were found with
two horni eggs, and only one had three. When more than one egg had been
placed on a given sucker, there was usually no way of knowing from their ap-
pearance if they had been deposited at about the same time or at different times.
Unlike the Beechwood area, the egg-bearing suckers in the Phelps and Kenton
areas usually had more than one egg (average 2.5) and one had a surprising
total of 15 (table 5). When the difference between the latter areas and the
Beechwood area are considered, it is only reasonable to conclude that the oc-

Carlson and Knight: Agrilus beetles 49

Table 5. The occurrence of Agrilus horni eggs in the Kenton and Phelps areas
on Populus tremuloides root suckers in height classes of 1 to 7 ft.

Number of Eggs
Average per

Suckers Bearing Eggs

Class
(feet) Total | Total | Egg-Bearing

Sucker

1 1 1.0

2 12 Lea

3 41 4.5

4 30 ag |

5 28 ve

6 14 Lye

7 8 LA

All oe Os

currence of more than one egg on a single sucker is usually not an indication
that they were laid in sequence by the same female. Instead, what seems to
be indicated is that some suckers were far more suitable for horni oviposition
than others, even of the same size. Moreover, in the Phelps and Kenton areas
the horni habitat was restricted to a very narrow zone, and movement along it
probably resulted in these beetles contacting suckers in particular locations
more frequently than others. The most reasonable conclusion seems to be
that when suckers in these particular locations were otherwise favorable for
oviposition there were a number of eggs laid on them.

Sucker size should be an important factor affecting the distribution of eggs
by horni females and/or the survival of the larvae. In both height and DAB
chi-square comparisons the difference between the observed and expected dis-
tribution of eggs is highly significant (table 6). The relationship of egg occur-
rence to DAB is more sensitive than that with sucker height (figs. 23-26).

The Kenton trend line of O,/E,, over DAB (fig. 24) shows a double peak,
but the second peak is reduced in the graph of S/E,, over DAB (fig. 26). This
suggests that in the Phelps area the suckers most likely to elicit horni ovi-
position had somewhat larger diameters than those in the Kenton area. This
appears to be true despite the fact that the abundance of small suckers rela-
tive to those of larger sizes was greater in the Phelps area. Examination of
the data on egg placement for the two areas yields the most likely reason for
this difference. In the Phelps area more eggs were located in association with
moss than on other kinds of deposition sites. Most of those Phelps area eggs
deposited in association with moss occurred on the 0.6 and 0.7 inch suckers
(all but 2 out of 21), and these two DAB classes were the source of the great-
est difference between the areas. Therefore, it seems very likely that the
occurrence of mosses on these suckers in the Phelps area is responsible for
the positive skewing of the trend lines (figs. 24, 26) as compared to the more
negatively skewed ones for the Kenton area.

Inasmuch as factors independent of sucker size, such as the occurrence of
mosses, affect horni oviposition, even two areas with nearly identical arrays
of sucker sizes will not necessarily have similar frequency distributions for
those bearing eggs. In general, however, the data do strongly indicate this
much: suckers between 0.3 and 0.9 inches DAB are most likely to elicit ovi-
position.

‘Jaaoy AjITIGQeqord [00 94} 4B JUBOTFTUSIS AT[VOTISTIEIS
* >

© ‘ZUISSIUL O18 SONTVA {YC OA} osnVdeq UOSTIedUIOD YSIoY OJ yey} WOTf SAFJIP [VIOL |
rf
& 489 ° 09 00 ‘FZ bz een
o Z9°ST LY ‘0 € 6IL‘°0< L
r e6 *eT LL‘0 7 6TL°O - Z19°0 9
= OFS 92 “T v TT9°0 - POS "0 G
= 8S ‘OT (1) ae 8 €0S°0 - 96€ °0 v
> OT'0 6S°E € G6E°0 - 882 °0 e
66°S 0g °9 “4 L8Z°0 - O81 ‘0 Z
Q 0S ‘IT 0S ‘TT 0 I sdjeud
% axf0 CT. 00°82 8Z Eco mem
= eL°0 €L°0 0 i L
A €% 0 Zert z 6S6°0 - 608°0 ¢ 9
H 61°9 97 °Z 9 808°0 - §S9°0 v G
= LS ‘0 LG" G ZS9°0 - 90G°0 6 7
< vS'P 91 °S OT G0S‘°0 - 6SE ‘0 G g
2 G12 €8'9 € 8S$‘0 - 222 ‘0 p rd
ee 29°F SL's rd IIZ"0 > I I uojUe yy
g ee &. Cogs Ss) (“O) (soyout) Coz xta), « <l@) (393)
'S) pojoodxy poAresqo sse[O poyoodxy peArosqo SSseIO BolV
sreyong Bul1eveg-334 qva Ssleyong Sulreeg-33q ISIoy

‘svore sdjoyg pue uoWoy oy} UI s83e TuIOY SN{lIsV SUlIveq S1oyONS Soplofnwie.s} snindog jo
suoljnqizjstp Aouenbeaj sse[o-9ezIs poJoodxo pue paArosqo jo uostzeduiod oazenbs-1y9 *9 eTqQeyL

20

ol

: Agrilus beetles

Carlson and Knight

‘Tyq Jeyons jo uotjounj & sv pojo[d ALoseyeo

(aseq ye TOJOWILIP) qyq IeyONS YoVe AOJ ssoyons SuLrveq

-23e jo (“q) saequinu pojoedxe poysn{[pe pu (“O) paAresqo
ay} Jo olyer :s830 TUAOY SNilasy jo uolnqiaysiqd “Pe “SIA

(seyoul ul) qyvd
O'l 6'0 8'0 Z0 9°0 S ‘0 ae) £0 Z0 ie) 0

sdjoud °
uojUey x

Me TO

‘yyStoy Aoyons Jo uoTjouny

® se popold A1080189 JySloy LayONS YORE IOJ Szoyons Surazvsq

-33q Jo (“q) szequinu poyoodxo poysn{pe pue (*O) poArasqo
ay} JO Olver :sS830 TUIOY Sn{lisy Jo uolnqiajsiq “EZ “BIA

(joo} Ul) JYSTOY
9 G 4 £ G Ll 0

sdjoyd e : i)

uojuey x
z 0
£0
v0
¢‘0
9°0
£0
80

60

we / 0

Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

D2

‘Eva Jeyons jo uotjouny & se poyjo[d pue A1080}e0 "yySIoy IayoNs Jo uoTJoUNy & SB po}

ged sieyons Zutreeq-38a jo ("“q) requinu poyoodxe pejsn[pe oy} = - 40d pur Rackeies zed szoyons Zurreeq-38e jo ("“q) szequinu po
Aq pepratp ‘AroZeyeo (eseq ye ToJoUIeIP) CVC yoRe Jo s1eyoNs -y0a0dxe poysn{pe oy} Aq poptatp ‘A108eze0 yystoy YORE Jo s1eyons
Zul reeq-33e uo punoj (g) s88o TUIOY SNTIISY [e1IOL “9% *STA Sutszeoq-380 uo punoj (g) s880 Tu10Yy SH[ISy [eI0L “Gz ‘81a
(soyoul ul) qv (Jo9} UT) JYySTOH
O'l 60 80 £0 990 ¢ ‘0 v0 £°0 (Aa) ie) 0 L 9 $ Vv € 4 Ll 0
sdjoud o
uojuey x
l l
2)
z i
sdjoud o a
uojuey x =
oo
fy
ex £
y y
¢ g

Carlson and Knight: Agrilus beetles D3

Without data on the DAB frequency distribution of a stand to compare with
that of its egg-bearing suckers, one probably would not come to the same con-
clusion. In other words, the information obtainable by sampling at random in
respect to sucker size and tallying only data pertinent to egg-bearing suckers
would be quite different from that presented above. Hypothetical numbers of
egg bearing suckers for Phelps and Kenton are plotted in figure 27a, and
represent what might have been expected from a random sample of 1000 suck-
ers in each area, as calculated by Carlson (1968). Figure 27b shows that 2/3
of the egg-bearing suckers in the Phelps area would probably have been below
0.4 inches DAB, while a like proportion for the Kenton area should have been
below 0.5 inches. Figure 27a agrees with a priori expectation for a random
sample in that the curves are negatively skewed and have less pronounced
maxima than those in figure 25. The negative skewedness of the former is
due to the preponderance of suckers of smaller sizes (the ‘‘j’’ age-frequency
distribution). A possible explantion for the disparity between the two curves
is that horni tends to oviposit on suckers that may be somewhat above optimum
size for survival of larvae as a natural compensation for the preponderance of
suckers which possibly are below optimum size. A study of the relationship
between sucker DAB, root size, and survival of horni larvae would, of course,
be necessary for proving this hypothesis.

The mechanism which restricts horni egg placement to suckers within
moderately narrow diameter limits is unknown, but it may be proprioceptive.
Proprioceptors that could operate in measuring suckers could be located in
the legs, antennae, or even the ovipositor. There are no obvious morpholo-
gical features differing from those of either anxius or g. liragus that-could be
singled out as possible sites for the supposed proprioceptors. However, hor-
ni varies much less in size than either anxius or g. liragus, and the genetic
basis for this lower variability may not be unrelated to horni host selection
behavior.

In reference to horni, Graham et al. (1963) say that ‘‘the root-girdling
Agrilus deposits its eggs on the bark either near the ground or sometimes a
foot higher.’’ Three inches above ground line is the maximum height at
which we have found horni eggs under natural conditions. Infact, we found
150 eggs below the 1 inch level, while encountering only 34 above it, a highly
significant statistical difference from the expected (table 7). All but 2 of the
latter (table 8) were below the 2 inch level. Therefore, Graham et al. (1963)
were probably mistaken in identifying anything at the 1 foot level as horni
eggs.

The horni eggs were commonly deposited on rough bark (lenticels, cracks,
flakes, etc.) against adventitious buds, against or beneath the leaves and stems
of mosses which sometimes grow on the bases of the suckers (table 8). Oc-

Table 7. Chi-square comparison of the number of
Agrilus horni eggs above and below the 1 inch level
on the bases of Populus tremuloides root suckers.

Height above
Ground
(inches)

Number of Eggs
Observed Expected

** Statistically significant at the 0.01 probability
level.

54

Egg-Bearing Suckers

Cumulative Number of Egg-Bearing Suckers

Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

iw) o1 ~]
o1 © O1

Expected per 1000

i=)

275 (b)

250 , pin

225 aa wert
x a

200 e

jt
~]
o1

.e)

_

eo)

=)
t=

Expected for a Sample of 1000
bo on
ee)
Ta

Se

X< Kenton

- © Phelps

o1
j=)

o. Gog Gos ee re Oe OR ee RE
DAB (in inches)

Fig. 27. Hypothetical distributions for Agrilus horni eggs: the num-
ber of egg-bearing suckers (a) and cumulative number (b) expected for
random samples of 1000 Populus tremuloides suckers for areas near
Kenton, Michigan (dashed line) and Phelps, Wisconsin (solid line).

Carlson and Knight: Agrilus beetles s}9)

Table 8. Position and situation of Agrilus horni eggs on Populus tremuloides:
pool of data for 1964 (Beechwood area) and 1965 (Kenton and Phelps areas).

Eggs Situated on or Against:
Smooth} Rough Bark
(flakes, etc.) | Buds Moss

- Height Above
Ground
(inches)

casionally, they were also found on smooth bark, but most of those eggs were
located at ground line where they had probably been partially supported by or-
ganic debri of some sort. A chi-square comparison between the height distri-
bution of eggs laid on smooth bark and that for all other eggs shows them to
differ significantly (table 9). While the number of eggs laid on smooth bark
above ground line is less than the expected value, the reverse is true for the

number observed at ground line.

Table 9. Chi-square comparison of positions of Agrilus horni eggs on smooth bark
with positions of other eggs (pooled data from Beechwood, Kenton, and Phelps areas).
Number of Eggs
Observed ___Expected?

Smooth
Bark

Height Above
Ground
(inches)

Other

tExpected = (observed column total + grand total) x row total.

*Statistically significant at the 0.05 probability level.
Mating Behavior

FUNCTIONAL MORPHOLOGY

Aedeagus. Before entering the discourse on the behavior of individuals
during courtship and mating, it will be necessary to describe some anatomical
structures important in mating. Basing the interpretation on Lindroth and
Palmén (1956), the aedeagus (fig. 28) of buprestid males is composed of an
eversible penis (pe), which is sheathed by the tegmen. The tegmen is com-
posed of the paired parameres (pa) and an unpaired basal piece (bp). The
junction between the basal piece and the parameres is not evident in Bupresti-
dae. Hence, the term parameres will be used to denote the part of the tegmen
which is divided, the part often called the lateral lobes in buprestid taxonomic
discourse. The latter, and the associated designation of the penis as the medi-

56 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Fig. 28. Dorsal side of the aedeagus of Agrilus granula-
tus liragus; pe = penis, pa = parameres, and bp = basal piece.

an lobe, are not useful terms for functional descriptions.

On the ventral’ side of the tegmen (figs. 5-8) of species in the anxius group,
there is a trough or groove which extends from near its base to the tips of the
parameres. It is very narrow at the base of the tegmen and gradually increases
in breadth toward the apex, broadening most in anxius, but much less in horni.
During coitus this groove receives the pygidial spine of the female, which serves
to guide the aedeagus into position. The groove also appears to have functions
connected with sexual recognition which are discussed later.

There is also a groove on the dorsal side of the aedeagus (fig. 28), although
it is rather shallow in comparison to that on the ventral side. The penis is
only slightly recessed below the dorsal surface of the parameres, which are
rather flattened, and the groove extends basad of the parameres, sometimes
nearly to the base of the tegmen.

Movements of the heavily sclerotized aedeagus appear to be controlled
mostly by the large muscles that are directly attached to it. The ovipositors of
buprestids, on the other hand, are not heavily sclerotized. As there are no
muscles directly attached to them, they are principally hydrostatic organs.

Female Organs of Reproduction. The only interpretation of the morphology
of a buprestid ovipositor known to us is that of Tanner (1927). Because the ovi-
positors of Buprestidae appear to differ from those of other Coleoptera, it seems
best to employ his interpretation. After the homologies are worked out, it may
be possible to use less burdensome terminology.

The ovipositor of species in the Agrilus anxius group (fig. 29) differs from
the configuration shown by Tanner (1927) for Chrysobothris debilis Leconte in
a single important way. The C. debilis ovipositor, which is similar to that of
a Buprestis sp., has the bacula of the proctiger and coxite separated by the
entire length of the valvifer baculum. Species in the A. anxius group have the
bacula of the proctiger and coxite separated only by the thickness of the valvi-
fer baculum. The valvifer baculum lies beside that of the coxite, and the two
are nearly parallel, except at their bases, where the valvifer baculum inter-
venes between the bacula of the coxite and proctiger.

Figure 29 shows the Agrilus anxius ovipositor in its various attitudes. A
comparison between fig. 29, c and d shows how the ovipositor can be likened
to a tube within a mostly membraneous sac. The part of the ovipositor bear-
in the coxites will be referred to as the ovipositor tube. It appears that only
this part of the ovipositor is actually extended from the abdominal cavity dur-
ing oviposition.

During copulation the male aedeagus is inserted to the point where the tip

Sreuia in the retracted position, as is traditional; not ventral sensu Lindroth and
Palmén (1956).

‘une 9}1X0O = qo pus ‘WIN[NoVq JOjIATVA = QA ‘UIN[Noeg TeST{00I1d = qq ‘Je AIS = 19

‘BANA = NA ‘ies1j00Id = Ig ‘wInjool = Y ‘g olUtojs = Bq *Q 0418.19} = 8r :sjoquiAs oy, “eplon] Brewed B Jo pre oy} JNoyWIM

UMBIP 9IOM SMOIA [[VY “UOT}ISOd pojOVIjOI Ul [e.10}e[-O1JUOA ST GQ UI UMOYS MOTA OUT, *ATeJETAUIOD Sev IOJISOdIAO oY} pUle}xe

0} 1vodde jou Op SeTeUley BUT}ISOdIAO Jng ‘SUOT}YeAJSNII[I oy} SOTI[dWIIs DO Ysno14} VW UI WMOYS UOTSUE}X9 JO 90IZ0p oy, °* (OTK
qnoqe) uoT}Isod pepuezxe Ul SnHIxue SN[LISy JO 10}{ISOdtAO 94} Jo syoodse (9) [esiop pue ‘(q) Teqquea ‘(y) TeteyeT “6% “STA

o7

Carlson and Knight: Agrilus beetles

+S Gd uy

ni dd ge

58 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

of the female’s pygidial spine rests in the basal end of the ventral groove on
the tegmen. When the aedeagus is in this external position, its probable in-
ternal position in relation to the various bacula of the ovipositor tube is as
shown in fig. 30. It is assumed that after this position is attained, the male
everts the penis into the vagina (Va, fig. 31). The tip of the penis probably
fits into the receptacle (Rp, fig. 31). The wall of the vagina has a complex,
thickened structure which gives it great elasticity. It is therefore assumed
that after the penis apex enters the receptacle, eversion of the penis contin-
ues until it is fully extended. It appears that the elasticity of the vagina may
allow successful interbreeding between even the individuals at opposite size
extremes in populations with extensive size variation.

After the penis is fully extended, the internal sac is evidently everted up-
ward into the bursa copulatrix (Bc, fig. 31). The minute spinules surround-
ing the opening to the bursa apparently guide the sac into place. They probably
also serve to hold the sac in place as insemination occurs.

BEHAVIOR OF MALES

Sexual Recognition. When a male (A)' mounts another individual (B), he
almost immediately exserts his aedeagus, which at first projects backward,
but then rotates downward and forward such that its apex contacts the tip of
the abdomen of individual B. After the abdomen of B is contacted, the subse-

Fig. 30. Probable position of the dorsal side of the aed-
eagus (solid line) relative to the bacula of the ovipositor
(dotted line) during copulation in Agrilus granulatus liragus
(cf figs. 28 & 29).

Fig. 31. Lateral view of the Agrilus anxius ? internal organs of reproduction (less
ovaries) and ovipositor (retracted position); Sp = spermatheca, Rp = receptacle for
penis apex, Be = bursa copulatrix, Sa = spiculate area of vagina, Co = common ovi-
duct, Va = vagina, Ty = tergite 8, Sg = sternite 8, and Vu = vulva. Drawn without
the aid of a camera lucida.

biienenthen courting males are referred to as ‘‘A’’ and the individual being courted as “‘B.’’

Carlson and Knight: Agrilus beetles 29

quent behavior of the courting male (A) depends upon the position of the hind
sternite of individual B. If the sternite is in lowered position, insertion and
copulation ensue, provided Bis afemale. If Bis a male, the tips of the aedea-
gi of both come into contact and male A dismounts quickly.

When the aedeagus dissected from a male was placed into the abdominal
opening of virgin females, courting males responded to them just as they did
to males, provided the aedeagus artificially implanted in the females had not
moved to the side of the abdominal opening. Therefore, it appears that the
responses of courting males are elicited by what they touch with the aedeagus.
The observations were few, however, and only one observation was made on
a male courting another whose aedeagus was removed. In this case insertion
of the aedeagus occurred, and when the behavior of male A indicated that he
was everting the internal sac, he was removed, whereupon the sac was re-
vealed protruding from the tip of the penis.

When a courting male probes an individual not having the hind sternite in
lowered position, the process of sexual recognition is somewhat more intri-
cate. Before the behavior of courting males in this situation can be described,
some sexually dimorphic structures of the last abdominal segment must be
discussed. The role played by the female pygidial spine during insertion of
the aedeagus has already been partially considered. Were this the only function
of the spine, it would be difficult to explain its presence in males. Moreover,
male pygidial spines are often somewhat longer (relative to body length) than
those of females. The significance of this dimorphism seems to relate to
the function of other sexually dimorphic structures of the hind segment. In
both sexes the marginal sclerite of this segment is smooth and flattened, and
a groove separates it from the sternite. The apex of the hind marginal scler-
ite of females is angled slightly upward, or beveled, with respect to the ster-
nite (fig. 32). In males, however, its apex lies in almost the same structural
plane to which the sternite conforms (fig. 33). The groove between the mar-
ginal sclerite and the apical part of the hind sternite seems to be somewhat
larger in females than males, but this appears to be partly due to the differ-
ences in the angulation of the marginal sclerite.

Fig. 32. Abdominal apex of Agrilus Fig. 33. Abdominal apex of Agrilus
eranulatus liragus ; arrow points to the granulatus liragus ©; arrow points to the
groove between the marginal sclerite groove between the marginal sclerite

and sternite of the last segment. and sternite of the last segment.

60 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Let us return to the situation in which a male probes an individual whose
hind sternite is in the closed position. The tip of the aedeagus of A first touches
the marginal sclerite of B at or near the point where it closes against the py-
gidium. After this point is contacted, A draws the proximal portion of the
aedeagus upward. This causes its tip to move ventrally, since the pygidial
spine of B acts as a fulcrum, fitting in the ventral trough of the tegmen about
midway along the parameres. When the tip of the aedeagus moves downward
and forward it comes into contact with B’s sternite. When B is female, the
sternite resists the downward movement of the aedeagus, as the latter catches
in the groove between the marginal sclerite and sternite. On the other hand,
if B is male, the groove fails to catch the tip of the aedeagus, which slips
freely past. This appears to be due to the combined features of the apical
marginal sclerite and the pygidial spine of male B.

When males are probed, dismounting is the only outcome. This is elicited
either when a probing male’s aedeagus slides past the groove between male
B’s marginal sclerite and sternite or when its aedeagus contacts that of B.
The probing of females, on the other hand, has several possible outcomes, de-
pending upon her receptivity. Immediate insertion of the aedeagus occurs when
B has her hind sternite already lowered prior to probing, but when it is not,
the prying movement of A’s aedeagus may lower it, provided B is a receptive
female. If she is not, a male may be unable to insert its genitalia, and will
persist in probing until (1) coitus is achieved, (2) he is knocked free of B as
a result of her ‘‘frenzied’’ scurrying, or (3) until he fatigues (or appears to
fatigue).

The hypothesis that sexual recognition by males is chiefly a physical pro-
cess was tested in the laboratory by altering the structure of the last seg-
ment in both sexes with glue. Some evidence was obtained in support of the
hypothesis, as the attempt to create a female-like male was partially success-
ful. Males behaved quite differently after mounting and probing altered males
than they did when courting normal males. All of the results obtained at first
did not seem consistent with the hypothesis, but it was discovered subsequent-
ly that consistent results cannot be expected unless the last segment of an al-
tered male is held tightly closed. After this was taken into account, the re-
sults were less variable. Some males then mounted and probed altered ones
several times in succession before dismounting. Others continued probing
for up to 5 minutes.

Physical contact definitely plays an integral part in sexual recognition by
males. The important question that remains unanswered seems to be this:
when a female holds the hind sternite in closed position as she is mounted and
probed by a male, is the response of the male dependent entirely upon physi-
cal contact or does he receive some additional stimuli (e.g. chemical) which
work in combination with the physical ones? If attempts to modify the struc-
ture of the hind sternite of females had not failed completely, a more definite
answer to this question might have been possible. Unfortunately, the minute
size of the crucial structures makes it impossible to smooth them over with
glue or similar substances, like those used to modify the structure of the hind
sternite in males. The mass of the applied material inevitably produces a
bump which is more of an obstruction to the aedeagus than the groove between
a female’s hind marginal sclerite and sternite. Other means must be found to
smooth over this groove and there must be more tests carried out with females
that have an excised male aedeagus implanted in the opening to the abdominal
cavity. The small amount of evidence obtained by the latter technique seems

Carlson and Knight: Agrilus beetles 61

to indicate that males may not recognize females by means alternative or ad-
ditional to the physical ones.

Insertion of Aedeagus and Insemination. After insertion of the aedeagus
is initiated the female pygidial spine apparently continues to play a significant
role for successful coition. It continues to act as a fulcrum while the male
alternately lifts the proximal part of the tegmen and then pushes the aedeagus
into the ovipositor tube (fig. 34, 39c). These movements by the male seem to
assist insertion by putting pressure against the female’s hind sternite to push
it downward, and the sheath-like lower part of the ovipositor tube is probably
simultaneously pulled away from the fleshy upper portion.

When full insertion of the aedeagus is achieved (fig. 39d), the male’s hind
sternite is nested in the notch between the tips of the female’s elytra, while
her pygidial spine rests in the proximal part of the ventral trough of the aedea-

Fig. 34. Mating of an A. anxius & and an A. granulatus liragus &.
The aedeagus is being pushed into the ovipositor tube (not visible) foll-
owing the prying of the last sternite away from the pygidium. This
downward prying occurs when the male raises the abdomen, which
brings the median part of the aedeagus into contact with the pygidial
spine of the female, which acts as a fulcrum.

gus (which faces dorsally in the inserted position). After several minutes the
male begins a series of vibrant movements, at which time the middle and hind
legs are usually extended laterally. Eversion of the internal sac probably
starts with commencement of these movements, because males removed from
females during this phase of coition had the internal sac partly everted, pro-
truding from the gonopore. Males removed somewhat later lose the internal
sac which is apparently restrained by the spinules around the opening to the
bursa.

Behavior of Males Under Natural Conditions. Information about Agrilus spp.
suggests that they depend upon host selection, at least in part, to bring about
the meeting of the sexes. In other words, the kind of site utilized for mating
by a particular species depends on many of the same factors which govern the
selection of sites for oviposition. Therefore, these are also the factors which

62 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

prevent interspecific matings from occurring, while making it quite likely that
intraspecific pairings do occur.

Bronze Birch and Bronze Poplar Borers. Excepting the
host specificity factors, the mating behavior of anxius and g. liragus is simi-
lar. For efficiency in making observations, most field work on the mating
behavior of these species involved watching individuals which were on wind-
falls (fig. 35), logs, or logging slash (fig.36). It is fairly safe to assume that
they do not behave very differently on these hosts than on upright ones.

The males of g. liragus and anxius move along the stems of their hosts,
usually remaining still for only brief but frequent interludes. On very warm
days, the movements of these beetles are rather rapid and the hesitations are
very brief, so that they move rather continuously with a jerky motion. On
cooler days they remain stationary for rather long periods, and are not con-
spicuous. These insects are not active at temperatures below 70° F (Barter
1957, 1965).

When any arthropod passes within a foot or so of these Agrilus males, the
males approach the former. [If the other animal, in turn, approaches them
aggressively, such as was observed with ants and males of a larger buprestid,
Chrysobothris sp., the Agrilus male responds by releasing its grip on the host
and falling free of it. However, if the other arthropod is moving away from
the Agrilus, the latter pursues. In one instance a g. liragus male was seen
following a Dicerca (another buprestid), and upon overtaking the latter, the
g. liragus male actually scurried part way up onto the elytra of this much lar-
ger insect.

When two anxius or g. liragus males approach each other, they usually do
so more gradually, at least if each has detected the other, which usually is the
case. After the males stand facing each other from a distance of one to several
inches, one of them quickly approaches and mounts the other. Such homosexual
encounters occur very frequently in nature, much more frequently than hetero-
sexual ones. They may have a special significance, particularly for anxius

and g. liragus.

Homosexual encounters seem to have a survival value that is real. A. anx-
ius and g. liragus males exhibit a sort of quasi-territorialism on their hosts.
After a male mounts and probes another, he usually flies or walks to a different
part of the host. These encounters between males seem to have an interesting
uniqueness in that it is the aggressive male that is usually repelled. There is
nothing to indicate that the aggressor can be repelled by anything short of
mounting and probing. The seemingly inefficient means by which the males of
these Agrilus recognize each other may be important in keeping them spatially
separated. This quasi-territorialism among males may serve to increase the
probability of occurrence of heterosexual encounters.

The beneficial effects of homosexual encounters may not be derived with-
out some cost to the sexual responsiveness of the aggressive male. During
laboratory studies of behavior it was discovered that if males are removed
from females just after completing the sexual recognition phase of courtship
and are immediately released again, they often mount the next individual they
encounter (either sex). The sexuality of males is definitely enhanced when they
mount and probe females, and it is apparently not greatly diminished by arti-
ficial termination of courtship. However, males whose sexual responsiveness
has been enhanced in this way, seem to lose it after a short sequence of homo-
sexual encounters. Perhaps the decrease in sexual aggressiveness occurs as
an integral part of the reaction which causes an aggressive male to move away
from the one he has mounted.

Carlson and Knight: Agrilus beetles

Fig. 35. Populus tremuloides windfall, typical of those where
the adult behavior of Agrilus granulatus liragus was observed.

Fig. 36. Betula papyrifera logging slash, a typical observation site for the

studies of the adult behavior of Agrilus anxius.

63

64 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Apparently, there is some sort of threshold which must be reached be-
fore courtship is elicited in males, and when that threshold has been |
reached or exceeded, males court immediately upon encountering other indi-
viduals. All males on their hosts under natural conditions seem to have high |
sexual responsiveness. This indicates that factors associated with their attrac- ©
tion to mating sites and their pre-courtship behavior may induce this responsive-
ness.

During their vigil on the host, anxius and g. liragus males are often ob-
served repeatedly exserting the aedeagus. This behavior is also exhibited in
the laboratory by males not on their hosts, and it seems to occur well in ad-
vance of the attainment of the mating threshold. It could be a protagonistic
aspect in a part of the male behavioral repertoire that is somewhat self-regu-
lated as a balance between protagonistic and antagonistic aspects. The exist-
ence of this kind of behavioral balance has been demonstrated in other groups
of insects (see Kennedy and Booth, 1963; Graham, 1959, 1961).

Little has been established concerning the extent to which the energy of
females, once they have reached the host, is spent in activities which lead to
encounters with males. Most field observations seem to indicate that they are
rather passive. On the other hand, laboratory studies of mating indicate that
females may court males. Ona number of occasions, virgin g. liragus fe-
males mounted males and displayed many of the actions typically manifest by
males in courtship. In some cases the ovipositor was even exserted. Virgin
anxius females never showed such a striking ‘‘maleness’’, but occasionally
they moved part way up onto the elytra of males and flailed the antennae in the
manner of courting males. Whether or not virgin females would show the same
sort of response to other females is not known. No such behavior was ever
observed in the field, and must happen only rarely, if at all. Moreover, un-
der natural conditions, males seem to most readily mount individuals which
turn away from them. It is difficult to reconcile this with active courtship by
females, unless it is natural for the sexual aggressiveness of females to be )
subordinated by that of advancing males, with females terminating their court- _
ship by turning away from males. |

According to Barter (1965) g. liragus males ‘‘actively seek or ambush non-
ovipositing females.’’ He says that ovipositing females are unattractive to
males, but this does not appear to be an entirely accurate statement. Barter
observed mating less frequently between 1:00 and 4:00 PM, the peak hours of
oviposition. This alone does not show that ovipositing females are unattractive —
to males, however. Males simply do not encounter ovipositing females fre-
quently. Females ovipositing on logs are found along the sides of the logs on
days when the males orient more to the upper surface. Ovipositing females are,
therefore, usually not within the line of sight of males.

Because females are motionless while eggs are being deposited, males are
not likely to detect them then. However, ovipositing females are approached
by males which detect them moving between egg deposition sites. One ovi-
positing g. liragus female under observation was approached by two males.
After moving away from the first male, she was encountered by the second.
Although she also moved away from the latter, he overtook her and mounted.
The female stopped moving only after the male had inserted the aedeagus, af-
ter which copulation proceeded normally. Therefore, ovipositing females are
not unattractive per se, although they may be rather unreceptive to courting
males.

Laboratory observations indicate that females courted after having recent-

a a 9 ag a ep Pn pe en

Carlson and Knight: Agrilus beetles 65

ly mated are particularly unreceptive. When females are totally unreceptive,
courting males are usually unable to insert the aedeagus. It appears that total
unreceptiveness occurs only after insemination occurs. When males are re-
moved from females during the course of mating and prior to insemination,
the females are usually receptive to the next male to court them.

The Aspen Root Girdler. Although g. liragus and anxius are pro-
vided with a means for meeting of the sexes which apparently depends upon
odors coming from bio-chemical degradation of the host, horni mating is
brought about through more subtle means. A. horni males are apparently able
to select hosts of a favorable physiognomy and in such ecological situations
that the chances for encounters with females are enhanced, but it hardly seems
as certain a means as that common to anxius and g. liragus. There is a pos-
sibility that horni males or females emit air-borne sex attractants. There is
no evidence to suggest that this is the case, however. A. horni males merely
seem to orient to poplar suckers which females are likely to land upon. They
sit on the sucker stems, facing upward, and their position usually ranges from
near the ground to near the point where the sucker’s lowest side branches are
produced, a height of about one to three feet. The males stay in nearly the
same position for periods averaging about 15 minutes, during that spell usu-
ally moving only when another arthropod moves near it on the stem. After
such a period, they move upward and out onto the leaves to feed for one to
several minutes before resuming their vigil on the stem. One male remained
on the same sucker for more than 2 1/2 hours, but this may be an exception-
ally long stay.

The entire horni courtship and copulation was observed only once in the
field. In that case, as the female came down the sucker stem and neared the
stationary male, he, in turn, approached her. The female then flew, but
landed again on the foliage of the same sucker, while the male moved rapidly
to the base of the sucker and turned upward. The female came down the stem
again, and when she was approached by the male she turned upward. The male
moved up quickly from below and mounted. After they had been in copulo for
about 15 minutes, the female crawled up the stem and onto a leaf where the
male dismounted. In one other instance in the field, the last part of horni
copulation was seen, and it terminated just as in the preceding one. In each
case, the female flew to another sucker as soon as the male dismounted.

There is an interesting parallel between these field observations and
laboratory observations on horni copulation. In the laboratory, too, the males
remained mounted after the females began moving around. In one instance a
copulating female which had remained still for 10 minutes, ran about in the
petri dish for 24 minutes with the male remaining mounted. In another case,
the female remained placid for 17 minutes, but ran about for the remainder
of the mating period, which totalled 21 minutes. In a field cage (fig 37), a
copulating female which fell from an aspen sucker with the male still mounted,
ran about on the ground for 40 minutes before the male was dislodged. The
tegmen was partially to fully inserted during the entire 40 minutes. Several
times the female seemed to orient towards debris which she could pass be-
neath, but which would not allow the male to pass freely. It is difficult to
believe that the behavioral repertoire of horni females could include sucha
specialized facet for dislodging males. After it had occurred several times,
however, the possibility seemed less remote.

The males of anxius and g. liragus do not usually remain mounted long
after the females begin moving about. Barter (1957, 1965) found that the av-

66 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Fig. 37. One of the field cages where some observations
on Agrilus horni adults were made. Note that these obser-
vations are not those designated in the text as field obser-
vations; the latter, comprising the majority discussed, re-
fer to observations on beetles not in captivity.

erage time elapsed in copulo was 7 minutes for anxius and 5 minutes for g.
liragus. He says the maximum observed for anxius was 23 minutes.

FEMALE ACCEPTANCE OF MALES. Other species differences in mating be-
havior were discovered primarily through attempts to obtain interspecific
crosses in the laboratory. These Sl ane showed that the females of
Agrilus horni will not mate with g. liragus males. In 43 recorded attempts to
obtain this cross, none was successful, as the g. liragus males could not in-
sert their genitalia. Often, these males would remain mounted for extended
periods, up to 5 minutes or more, continually probing the hind segment of the
horni female.

~ In 29 recorded attempts to cross anxius females with g. liragus males, on-
ly 4 led to qimove A. pensus s females, on the other hand, appear to be re-
ceptive to g. liragus s males. They are also receptive to anxius males, but
may not be receptive to horni males. Ina number of attempts to pair ‘horni
males and pensus s females, the latter usually did not seem to be receptive, and
copulation resulted only twice. In one of those cases the pensus female did
not seem entirely receptive. The difficulties encountered with the survival of
pensus adults abbreviated the testing of this interspecific pairing, and left us
with inconclusive results. A. granulatus liragus females are promiscuous,
accepting males of any of the four species involved in these tests.

When a male mounts a female, he thrusts his head and prothorax forward
and flutters the antennae over her head and the base of her pronotum (fig. 38)!
and sometimes they are simultaneously drawn back rapidly making a long
stroke over the lateral parts of the pronotum (fig. 39). At the same time males
palpate the female’s pronotum near the depression lying just anterior to the
scutellum. Sometimes, males perform these antennating and palpating move-

The antennal movements during courtship appear to be the same as those occurring later.
Because it is extremely difficult to get the beetles into proper position for photographing
during courtship, I am forced to use photographs taken during copulation.

Carlson and Knight: Agrilus beetles 67

Fig. 38. Position of the antennae at one point during ‘*fluttering’’
by an Agrilus anxius “. The antennal fluttering observed during
courtship seems to be the same as that occurring during copulation.
Compare the position of the antennae with the positions shown in
figure 39.

ments(collectively, hereafter antennation) prior to probing with the aedeagus,
and when insertion does not occur quickly, they are repeated during the prob-
ing phase. Antennation also occurs as the aedeagus is being inserted; it is re-
peated occasionally until the male starts the vibrant movements thought to at-
tend insemination, and is performed following insemination as the aedeagus is
being withdrawn.

No incontestable evidence could be found regarding the nature of the inter-
sexual communication which leads to females accepting or rejecting males of
a particular species. However, it does seem to have something to do with an-
tennation by courting males.

If a female begins to move during insertion and the earlier phases of copu-
lation, antennation appears to tranquilize her again. Exceptions occur in some
interspecific pairings, or when females are unreceptive as a result of recent
copulation. In these cases females usually run about despite antennation by
males. Males may have difficulty withdrawing the internal sac and penis while
the female is moving. This may be the reason for the occurrence of antenna-
tion as the aedeagus is withdrawn, and the selective advantage which favored
the development of this behavior may stem from the reduction in incidence of
injury to the male genitalia.

A comparison of the courtship behavior of anxius, g. liragus, and horni
yielded some data concerning the basis for the unreceptivity of horni and anx-
ius females to g. liragus males. During antennation the males of horni and —
anxius usually contact the head and pronotum of the female. A. g. -liragus
males, on the other hand, often do not touch the female with the antennae dur-
ing courtship, and when they do, seem to do so very lightly. However, there
seems to be too much intraspecific variation for this alone to account for the
entire lack of receptivity of horni females for g. liragus males. Even when
horni males seem to antennate very tet they are accepted by conspecific
females, and the few times when g. liragus males appeared to antennate like
horni males they still failed to elicit receptivity in horni females.

~ Examination of the scape and the antennal sockets of these insects at high

68

Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Fig. 39. Positions of the antennae during the stroking movements
of males (here Agrilus granulatus liragus) in copulo or courtship, ar-
ranged sequentially: (A) being moved forward, (B) maximum anterior
extension, (C) beginning of backward stroke, and (D) finish of back-
ward stroke.

Carlson and Knight: Agrilus beetles 69

power (90X) indicates a remote possibility for high frequency vibrations or
slight sounds being produced by some of the movements involved in antenna-
tion. The undersides of the scapes (reposed) are flattened and striate. Simi-
lar striations are found on the lateral rim of the antennal sockets and on the
part of the front between the antennal sockets and the eyes (fig. 40). Movement
of the scape over the striations on the antennal socket and front, as occurs
during antennation, must produce some friction and thereby vibrations or
sounds. The question is whether or not such vibrations or sounds could be in-
tense enough to serve in communication. When the head is thrust forward dur-
ing antennation (fig. 39, above), the increase in hydrostatic pressure should
be transmitted to the antennae. This, in turn, may increase the friction be-
tween the scape and the striate surfaces on the antennal sockets and front.

The frontal striations are definitely strongest in horni, weakest in g. lira-
gus, and of intermediate strength in anxius. Consequently, any sound or vibra-
tion produced might be somewhat species specific. In comparing the structures
with other species in the anxius group, we found that some of them, vittaticollis
for example, have even stronger frontal striations than horni. Perhaps these
species should be studied first in an effort to determine the significance of the
striations.

Stridulation

The possibility that other auditory, vibrational, or chemo-tactic communi-
cations might be responsible for species discrimination by females cannot be
discounted. Both sexes do produce stridulatory sounds that have been record-
ed. The method of stridulation is one apparently peculiar to Agrilus, and has
been reported for A. solieri Castelnau & Gory and A. cinctus Olivier (neither
sp. in anxius group) by Schaefer (1949). It involves a rapid rotation of the
head in its prothoracic socket. Males of anxius, pensus, g. liragus,or horni
have not been seen making these movements during courtship. Moreover, the

Fig. 40. Frontal view of Agrilus vittaticollis ¢ (X42); within the
circle are the striate areas between the antennal socket and the eye
and on the underside of the scape. The striations are coarser in

this species than in horni, anxius, pensus, or g. liragus.

70 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

sounds produced by these stridulations seem to have no connection with female
recognition of males in these species. The only time during courtship that the
head is involved in pronounced movements is during antennation, after it has
been thrust out of its prothoracic socket. The only situations in which stridu-
lation has been noticed are ones which would indicate that the noises produced
could be classified as stress sounds.

In stridulating these beetles tilt the head slightly forward at the top. This
apparently brings the genal ridges! (the carinae at the lower edges of the gen-
ae; fig. 41) into position to contact the prosternal lobe. The genal ridges act
as the scrapers, while the fine imbrications on the prosternal lobe serve as
the file (fig. 42). The latitude of the rotational movements of the head during
stridulation is quite small, in the neighborhood of a millimeter. Because of
their rapidity it was difficult to count the movements, but there are approxi-
mately 5 head movements per second to each side of the center position. This
agrees roughly with audiospectrographs (fig. 43) which show that the file is con-
tacted about 10 times per second. The sounds produced are very soft. Con-
sequently, it was rather difficult to record them. !

As one might expect of sounds that are really only stress-induced, there is >
no indication that these Agrilus sounds are species-specific. These beetles
probably do not produce other kinds of sounds with this same stridulatory ap-
paratus. As previously noted, stridulation has not been observed in a situation
that would definitely indicate that it was not stress-induced. There were a few
indications that it may elicit repulsion of individuals that contact stridulating
beetles. Alexander, Moore, and Woodruff(1963) suggest that, inasmuch as the
sounds of Coleoptera are usually too soft to function at long distances, they
may function in several fashions among adults in close proximity. The stridu-

Fig. 41. Postero-ventral view of the head of an Agrilus horni °,
showing the genal ridge (X109). Note the fine striations which are
barely visible at one point near the posterior end of the ridge; they
may be important in stridulation, but the sharp posterior portion of
the ridge’s outer edge may also contact the prosternal lobe (fig. 42).

Ane, G. B. Vogt, U.S. Dept. Agr. , Ac R.S., Entomology Research Division, Washing-
ton, D.C. believes that ‘‘genal ridge’’ is the best possible term for this structure (per -

sonal communication).

Carlson and Knight: Agrilus beetles 71

Fig. 42. Finely imbricate interior surface (head is excised) of the
right half of the prosternal lobe of an Agrilus horni ? (X82).

latory apparatus of the species which were examined thoroughly a 2.
liragus, pensus, and horni) was not sexually dimorphic, and recordings indi-
cate that there is no sexual difference in the sounds produced. They do not
appear to be directly associated with courtship. Although unreceptive females
may occasionally stridulate, their stridulation does not seem to have any ef-
fect upon courting males. In homosexual encounters, a sexually aggressive
male is usually repulsed only after he mounts and probes the other, and stri-
dulation does not seem to be involved. Nevertheless, stridulation may function
in some other situations to prevent over-congregation of males or females, or
they may serve to repel individuals of unrelated Agrilus species which may
live on the same host. These stridulations should evolve toward nonspecificity
in the eliciting stimulus (Alexander, et al., 1963). It is, therefore, difficult,
if not downright impossible, to assess their significance by studying them un-
der artificial conditions. On the other hand, as these Agrilus stridulations
cannot be heard, except through amplification or by placing the insects next to
the ear, and because the movements cannot be seen at distances of more than
a foot or So, anyone endeavoring to determine their real function under natural
conditions, will, to say the least, be faced with a real challenge.

EVOLUTION

Phylogeny

When our work on Agrilus was initiated, we had not ils ae seriously e-
nough the possibility that anxius, pensus, horni, and g. liragus are not all
closely related. It seemed that the important q questions to be answered con-
cerned isolating mechanisms and clues to how these species evolved from a
hypothesized common ancestor. Because of this improper assessment, more
species than originally anticipated will be brought into discussions about phylo-

geny.

THEORETIC BASES. It is quite usual to regard some characters as more
reliable than others as indicators of phyletic relationships. Nevertheless, we

Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

12

*poods

IOJOUL MOT pues ‘1vds8 YSIy ‘yeTZ odvys ‘YIM pueq OpIM ‘Q'S UIVS [OAD] YIVWI :OTOM SSUI}OS TOZATIAQGIA oy, ‘eTqrssod
SB Yep SB SSUL{IVW oY} OYBUL OJ posn 919M SUNI OM], ‘“SUIPLOOOL [VUTSIIO OY} Jo poods ode4 ey} F/T ye ‘eAoqe pouoTjUoW
Ieytjdure oy} Aq Ul pes sTeusis yyM ‘azezATIAqiA Aueduiog o1sjOo Ty ACY B YIM opewM sem YdersorjOedsoIpne oy, *“1ep100

-~91 94} JO UOTyeIOdO Y}IM Po}BIOOSS¥ SpUNOS WLOI] DOUBIOJAIJUI SUIZIWITUIW ‘1apIODeI 9Y} WOIj VOUBISIP B ye pooV{d oq
0} DUOYMOIOIW OY} POMOTI[V STUT, “Tep10001 oY} poyeredo OYM ‘1910 9} 0} STSUSIS OAVS puv DUOYdOAOIW OY} ABA SoeTJOeq
94} Pley euO - ‘SSUIP1OOEI BSULYVU IOJ portnbeit 919M suOsaod OM], ‘“‘puodes red SaYyoUT ct Jo poods odey & 48 ‘ouoydoro Tu

V-€ed [OPOW UBoILOULY Ue pUe ‘LOTJI[dWIe ~P-E9Ld [EPO PAODOUSeYI B ‘TAPIODIAL ZHZVEILd [OPOW P1lOoVOUsSeY, B YIM
OpeUl SBA SUIPIODE.I ode} OUT, ‘9 SNSBIT] SMB[NUCIS SN[lIsV ue Jo spunos AiOVe[NpPIAIs Jo yderSor1jJOodsoIpny “EP “SIA

Carlson and Knight: Agrilus beetles 73

shall set forth the assumptions concerning those we use, and give our reasons
for not considering others to have the value accorded them by other students.

Let it be made clear at the onset, that we do not regard any characters as
being ‘‘nonadaptive.’’ Many systematists have used the term nonadaptive in
reference to stabilized characters, while referring to those subject to much
variation as ‘‘adaptive.’’ The irrationality of the idea of adaptivity and nonadap-
tive is discussed by Dobzhansky (1956), who maintains that all traits, or at
least the vast majority of them, are adaptive or have been adaptive at some
time in the history of their development. To be sure, some traits are more
stable than others, but their stability should relate to the kinds of functions ra-
ther than to the existence of functions versus the lack of them.

In many cases it is difficult to perceive functions for some seemingly trivial
traits, but this is a poor reason for assuming they do not exist. So it is with
the pygidial spine of species in the Agrilus anxius group. Without searching
for its function, one might take what seems to be the prevailing attitude in such
cases--that its stability indicates that there is no selection against it, probably
because it is of no great disadvantage to its possessor. Quite to the contrary,
it seems that the stability of the pygidial spine stems from its importance in
the process of sexual recognition. Any large changes in the form or size of
the pygidial spine calls for corresponding changes in the form of the male aedea-
gus and other structures. In genetic terms, it probably owes its stability to an
association with pleiotropic genes whose total influence on the developmental
system is spared extensive changes through the moderating influences of homeo-
stasis or coadaptation (see Dobzhansky, 1956).

The pygidial spine probably has at least as much stability as characters
ordinarily given tribal or generic value in other insect groups. Perhaps Oben-
berger was the one taxonomist who should have had sufficient worldwide famil-
iarity with Agrilus to recognize this, but he was convinced of the impossibility
of even dividing Agrilus into acceptable subgenera (see Obenberger, 1959).
What Obenberger (1957) implies is that, to be valid, each proposed subgenus
must have as its basis a single character which occurs among no species of
other groups. If any Agrilus character could meet this absurd requirement, it
would probably be the pygidial spine. Obenberger (1957) denies this possibility
because (1) a pygidial spine is found in the Palearctic forms Agrilus guerini
Lacordaire, A. fleischeri Obenberger, and A. ater L., but is absent in an al-
leged intimate ally, A. suensoni Obenberger; (2) its presence among about 30
African species makes it self-evident that it occurs among entirely unrelated
groups.

To counter the first point one need only ask what criteria he uses to ally
suensoni with ater and the rest. He offers none, and almost any should be more
subject to convergence than the pygidial spine is to total disappearance. Oben-
berger’s second point is even more indefensible because the presence of the
spine among African species cannot, in itself, demonstrate that it has arisen
repeatedly. There is no 4 priori reason for believing that even the Ethiopian
species he mentions do not share common ancestry with Palearctic ones. After
all, the Palearctic region does extend into northern Africa.

INTERPRETATION. Fisher (1928) has given the form of the tarsal claws pre-
cedence over the pygidial spine in his classification. Asa result, three spec-
ies which are here placed in the anxius group are classified quite apart from it
by Fisher. These are Agrilus ruficollis, A. champlaini Frost, and A. browni
(= lateralis, sensu Fisher [1928]; see p. 6). The bifid claws of these species

74 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

have the inner lobes bent toward each other. This trait appears to be the only
one which they have in common with the species Fisher places them near. The
pygidial spine, the marginal sclerite of the hind segment, and the general form
of these insects links them with species in the anxius group.

On the other hand, ruficollis, browni, and champlaini have more in common
than just similar tarsal claws. They resemble each other in the structure of
the pronotum, elytra, and male or female genetalia. In addition, ruficollis and
browni are similar in coloration. Their hosts are not closely related, but the
larval habits of all three species are somewhat alike. The larvae of ruficollis
bore in the stems of brambles (Rubus spp.). During the early stages of devel-
opment they often (cf Smith, 1892) form galls by making a tight spiral in the
phloem, while late-instar larvae move rather directly up and down the pith. A.
champlaini forms galls on ironwood (Ostrya virginia) and blue beech (Carpinus
caroliniana), but the late-instar larvae spiral mostly in the wood within the gall
which forms as the result of the phloem blockage caused by the gallery of the
early instars. Fisher (1922) reared browni from bayberry (Myrica pensylvan-
ica), and Knull (1922, 1950) says Myrica asplenifolia is also a host. The browni
larvae make no galls. According to Fisher (1922), after passing the winter in
the roots, the browni larvae spiral upward in the limbs for a considerable dis-
tance before they reach maturity.

The number and variety of similarities between ruficollis, browni, andcham-
plaini makes it unlikely that they could all be convergent. However, these spec-
ies differ in ways that seem to push their divergence to a rather remote era.
For example, it does not seem likely that a species living on Ostrya and Carp-
inus could be intimately allied with species whose hosts are Rubus spp. or Myr-
ica spp. Consequently, in figure 44, champlaini is shown as also having remote
connections with species that live on Fagaceae, which includes Agrilus bilinea-
tus, A. acutipennis Mannerheim, and a number of other forms. While cham-
plaini might instead be related to the line represented by anxius and its allies,
it appears to resemble bilineatus and related species in a greater number of
minor respects. For instance, the latter have the inner lobes of the tarsal
claws turned inward, although to a lesser extent than does champlaini.

Allies of horni. A. browni seems to be more closely related to horni than
to ruficollis (fig. 44). While its tarsal claws are like those of ruficollis, browni
resembles horni in having the vertex less strongly impressed, the front flattened,
and the marginal sclerites of the abdomen clothed with white pubescence. The
browni aedeagus is intermediate in form to those of ruficollis and horni, but
browni has a trace of the prehumeral carinae, which are completely lacking in
ruficollis.

While the eggs of ruficollis are placed at the axils of the branches (Hutch-
ings, 1922b) or leaves (Smith, 1891) of its hosts, those of browni and horni are
placed at the bases of the host plants. The larval habits of browni and horni are
very similar (excluding host differences). In each case the larvae bore into the
roots and over-winter there. The only difference is that horni bores farther out
into the root and spirals back to the stem, while browni spirals mostly in the
branches. The early-instar ruficollis larvae, on the other hand, over-winter
in the stem and probably spiral only during the time prior to this initial over-
wintering period.

Agrilus vittaticollis has the vertex strongly impressed as in ruficollis.
Fisher (1928) says vittaticollis is allied with benjamini Fisher and audax Horn.
The host of benjamini is unknown, but audax infests the upper branches of Ulmus

‘sotoeds Sulyew-|[[es sopnjout orty

1D

an

“ure}19e0uN ST AWoTjIoeds jsoy poyeusisep oy} YOTYM TOT sotoods OSOY} JO SOUBU OY} SMOTIOJ Yew uOTsSonb vy
YST194Se UB YIM poyTeUl ore susuITOeds OU Uses OABY STOYINE oY} YOTYA Jo Vxey jo soweu oyy, ‘*dnor3 SniKe STI
-T18V 9Y} UI pooe{d sulzoy onoaRoTeg pue OTJOLBON PEqILOSoep oY} JO oUOS AOJ 90.14 OYeTAYd [eoneu.OdAY a “Sly

IO oe Tlewg jo sze10g|

Carlson and Knight: Agrilus beetles

eee

juriceki *

fleischeri fleischeri *
fleischeri kurosawai *
opuli

satoi (?) *
anulatus liragus

semiaurovittatus (?) *
pranulatus granulatus

bilineatus
criddlei
tamanukii *

acutipennis
olivaceoniger
& other forms
ruficollis
niveiventris
fulminans

ater ater

vittaticollis
uadriguttatus

amawakii (?) *
tscherepanovi *

ater
subcuneiformis (?) *
ranulatus

| fleischeri coreicus *
fleischeri nipponicola *

76 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

rubra (Knull, 1934). <A. vittaticollis, however, has a root boring habit quite
like that of horni, but in its morphology vittaticollis bears more similarity to
ruficollis. Hence, vittaticollis is shown (fig. 44) as sharing common ancestry
with ruficollis, browni, and horni, but splitting from them before they diverged
from each other.

Without exception, horni has always been classified as a close ally of anx-
ius, on the basis of coloration, the structure of the claws, and the form of the
prehumeral carinae. The superficial resemblance of these species is, in fact,
so close that without a knowledge of their biology this alliance would seem most
probable. It is almost axiomatic that coloration is an unreliable indicator of
phyletic relationships. The amount of variation which the prehumeral carinae
show throughout the genus, and indeed their complete absence in numerous spec-
ies, even within the anxius group, suggests that they are also unreliable and
_ghould be used for classification only with particular caution. The morphology
of the tarsal claws appears to be of a similar nature. Nothing more than wild
speculations could be offered for explaining why these characters should be
variable among species, but relatively constant within. Nevertheless, there
probably are sound and logical reasons for it being so.» 7

Probably the most significant morphological features which appear to link
horni with browni, rather than anxius, are found in the structure of the male
and female genetalia. In browni and horni, the ovipositor is half as long as
that of anxius (relative to total body size). In correspondence with the female
genetalia, the aedeagus in browni and horni is shorter than that of anxius.

There are a number of other minor characters which seem to separate horni
from anxius, g. liragus, or pensus and show its relationship to browni. (1) In
horni and browni the marginal sclerites of all abdominal segments, except the
last, tend to be rather densely pubescent, but in anxius, g. liragus, or pensus
the second marginal sclerite tends to be glabrous. (2) The striations on the an-
tennal scape and on the front, between the antennal socket and eye, are nearly
as strong in horni as in browni, while they are weaker in anxius and, particu-
larly, g. liragus. (3) The elytra of horni and browni tend to be evenly sculp-
tured and rather evenly clothed with pubescence that is uniformly short. Those
of anxius, g. liragus, and pensus are not sculptured as evenly, having a longi-
tudinal costa and a depression mesad of the costa and between the basal third
and apical half of the disc. Frequently in g. liragus, and less often in anxius
and pensus, this depression is pruinose. The hairs on the elytra of the latter
three species are not uniform in length, nor are they evenly distributed. They
are sometimes very closely spaced near the suture, about halfway between the
midpoint and apices, particularly in pensus.

A. horni also resembles browni in a number of non-adult characters, that
are different in anxius and its allies. According to Fisher (1922), the eggs of
browni turn black before hatching as do those of horni. Those of vittaticollis
darken in the same manner (Brooks, 1913). The eggs of anxius and g. liragus
turn yellow or light brown prior to hatching. Barter (1957) says that anxius
eggs are covered with a whitish liquid by the female after a group of them has
been laid. None of the various workers who have observed horni, browni, and
vittaticollis have mentioned that such a substance was applied to the eggs of
those species. However, according to Mundinger (1941), ruficollis sometimes
covers its eggs with such a substance, but often leaves them nearly naked of
the covering.

After the first instar larvae of horni, browni, and vittaticollis bore out
of the egg, they go directly to the roots of their hosts. It appears that the lar-

Carlson and Knight: Agrilus beetles 77

vae of these species over-winter in the first instar, and generally bore much
farther than anxius or g. liragus before undergoing the first molt. The latter
species usually molt to the second instar soon after they enter the phloem,

and their pupal chamber is often formed during the same year that the eggs are
laid. Even when first instar horni larvae were introduced into aspen logs,

they bored for long distances without molting before succumbing’. Apparently,
for the larvae of horni, browni, and vittaticollis, the first molt is elicited by a
very different set of conditions than are necessary for its occurrence in anxius
or g. liragus.

soe daapand at least, the most significant way horni differs from anxius
or g. liragus is in its mode of host selection. A. anxius and g. liragus live on
decrepit hosts, and seem to be attracted to them by odors given off as a con-
sequence of the biochemical deterioration associated with decrepitude. On the
other hand, horni seems to depend upon thigmotactic means for distinguishing
poplar suckers of suitable sizes from unsuitable ones which are apparently no
more or less vigorous.

Hybrid crosses between anxius, g. liragus, and horni were made, but do
not seem to substantiate other evidence concerning their phyletic relationships.
The following crosses yielded viable eggs: o horni X anxius 9, and the recipro-
cal; o horni X g. liragus 97; and ig. liragus X anxius 9 (no reciprocal eggs).

First instar larvae of all crosses yielding viable eggs were introduced into
aspen and birch logs. Some larvae of all hybrids developed at least partially.
However, late-instar galleries were formed by & anxius X g. liragus 2, and
one & horni X g. liragus 9 imago emerged. This one hybrid beetle, a female,
has characteristics of both parental species. The ovipositor is of the normal
length for g. liragus, and the elytra bear a pair of faint efflorescent spots. On
the other hand, the marginal sclerites of the abdomen are uniformly pubescent
and the pronotum has a distinct cupreous tinge, which are horni traits. This
hybrid was fully receptive to horni males, but unreceptive to g. liragus males.
The only abnormality noticed was a possible deformity in the articulation of the
tarsal claws. However, it is not absolutely certain that their condition is real-
ly abnormal because the claws were not examined before the hybrid died and
had dried out.

If the conclusions regarding the phyletic positions of horni and g. liragus
-had been drawn before the hybridization tests were performed, inviability of
the eggs would have been expected. The successful rearing of a & horni X g.
liragus ? adult is a rather baffling perplexity. However, the success of this
cross seems no more phenomenal in view of its wideness per se than it is in
light of the known differences in the chromosome numbers of horni and g. lira-
us.
ae It is possible that A. olivaceoniger Fisher is related to browni and horni.
Although olivaceoniger r lacks prehumeral carinae, browni has only a trace of
them. The form of the olivaceoniger tarsal claws is intermediate between the
_browni and horniforms. In most other features, all three are very similar.
Nothing certain is known of the biology of olivaceoniger. Blanchard (1889), call-
ing it the olivaceous variety of acutipennis, said it occurred on poplar sprouts in

f The work of Engel (1968) seems to indicate that the behavior of horni larvae can be sim-
ilar to that of the larvae of g. liragus or anxius, depending upon host condition.

ni reciprocal of this cross could not be assessed because no horni females were recep-
tive to g. liragus males.

78 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Massachusetts. Frost (1912), on the other hand, claimed to have encountered
it quite often on oak, which would indicate that it may really be related to acu-

tipennis.
Fisher (1928) places olivaceoniger near criddlei and acutipennis. While
acutipennis appears to be related to bilineatus and allies (fig. 44) and lives on

Quercus spp., criddlei lives on willows and is related to the Salix borer group.
However, criddlei is not placed in the latter group because it has traits which
also indicate a relationship to horni and its allies.

Because criddlei is a gall-former, its mode of host selection may be simi-
lar to that of horni. The habits of the larvae are more like those of horni than
those of species in the Salix borer group. During the earlier stages of devel-
opment, the criddlei larvae spiral in the phloem, but they later feed mostly
within the xylem (cf Wong and McCleod, 1965).

A. criddlei resembles horni in having a deep, smooth eset groove on
the first and second abdominal sternites of males. Of the forms in the Salix
borer group, quadriguttatus quadriguttatus has the strongest abdominal groove,
but it is not as well-developed as in criddlei. This groove is weakly developed
in males of q. niveiventris and q. fulminans, but absent in A. guerini Lacor-
daire, a Eurasian species (see Stepanov, 1954).

The male and female genitalia of criddlei are very similar to those of horni,
although the apex of the criddlei penis is acuminate (fig. 45) instead of rounded
as inhorni. The penis apex of quadriguttatus and subspecies is acuminate like
that of criddlei. The elytra of criddlei and those of species in the Salix borer
group are also very similar. In addition to having the tendency for the forma-
tion of discal spots of pubescence, the elytra of all of these species are angu-
late at the apex (fig. 46a, b). This latter feature of the elytra appears to have
reached its maximum development in guerini, which has greatly prolonged ely-
tral apices with a very deep notch between them (fig. 46b).

The Nearctic Salix borers have well-developed prehumeral carinae. These
carinae are not strongly developed in criddlei and are only rudimentary in guer-
ini. The evidence suggests that criddlei may be a rather direct descendent of
the form ancestral to the triad of groups composed of the Salix borers, Populus
borers, and Alnus-Betula borers.

The Salix Borers. Of this latter triad, the Salix borers seem to be the least
specialized. According to several accounts of its larval habits, guerini usually
bores in the upper branches of willows (Gaudin, 1921, Lecomte, 1925; Horion,
1955; Lundberg, 1962), and its larval habits are quite analogous to those of
pensus. The larval habits of the other species of Salix borers are essentially
unknown. The size of the willows which serve as hosts of q. niveiventris and
q. fulminans (e.g. Salix lasiandra and S. lasiolepis) suggests that these Agrilus
may have larval habits quite similar to those of guerini. It does not seem likely ~
that the larval habits of nominal quadriguttatus would differ markedly from those |
of its other subspecies, but theformer hasas its host the black willow (S. nigra a)|,
a very large tree at maturity. Therefore, q. quadriguttatus could have habits
analogous to those of anxius or g. liragus.

Agrilus yamawakii Kurosawa is a Japanese species (Honshu, Shikoku, Kyu-
shu), which shows some resemblance to the Nearctic Salix borers. The figure
in Kurosawa (1957) shows a rather unusual undulate band subapically on the ely-

on reference to q. quadriguttatus Fisher (1928) said: ‘‘This is probably the species re- |
corded as having been collected on partly dead alders (Alnus sp.), although no specimens |
have been examined which were collected on that plant.’’ This is an obvious reference to :
the statement of Blanchard (1889) regarding the series from which Horn (1891) described

pensus (granulatus, sensu Blanchard [1889] ).

Carlson and Knight: Agrilus beetles

Fig. 45. Penis apex of Agrilus criddlei.

Fig. 46. Elytra of Agrilus ater (A), A. guerini (B), and A. quadriguttatus
niveiventris (C). Figures of ater and guerini redrawn from Théry (1942).

79

80 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

tra of yamawakii, and there is no mention in its description of a median basal
groove on the abdomen of males. However, its habitus is otherwise so much
like that of quadriguttatus and subspecies that it is included as a possible mem-
ber of the Salix borer group. :

Agrilus guerini is the only Palearctic species in the anxius group that has
been reared from willows. On the basis of its discontinuous east-west dis-
tribution in Europe, Horion (1950, 1955) considers it to be a postglacial relict
of a tropical group. Théry (1942) comes to similar conclusions because of the
peculiar form of the elytra. To be sure, there are a number of tropical forms
with elytra similar to those of guerini (see Obenberger, 1957), but they do not
appear to be at all related to guerini as they lack the pygidial spine. The type
of A. nevadensis Horn, which is a synonym of q. niveiventris Horn, has elytra
prolonged somewhat like guerini, but those of guerini are longer and more
arcuate on the inner margin.

The Alnus-Betula Borers. The Populus borers and the Alnus- Betula borers
probably arose independently from a Salix boring ancestor. Moreover, it is
highly unlikely that the switch from willows to a host in the birch family could
have been made by a specialized form similar in habits to anxius or g. liragus.
There would, in theory, have been obstacles to its occurrence in this way.
Host selection of anxius and g. liragus seems to be strongly reliant upon air- |
borne by-products of biochemical degradation from hosts that are injured or
under severe physiological stress. Although an attempt will be made later to
discredit the assumption that anxius and g. liragus are closely allied and that |
there are no intermediate species, let it be assumed for now that the assump-
tion is valid. In addition, let it be assumed that anxius was derived from a
Populus borer, although the direction of the change is immaterial to the devel-
opment of the theoretical case'. The foregoing assumptions yield three anom-
alies. (1) The Populus borer ancestral to anxius would have to be preadapted
for detecting those chemical substances of birches that would indicate the equi-
valent of the conditions it was adapted for on Populus. (2) Implicit in the defini-
tion of pre-adaptation is the necessity that the foregoing occurred in the absence
of Betula spp., but the latter presently occur in most of the areas where Popu-
lus spp. are found. (3) The alternative to a hypothesis of preadaptation is for
the capacity to recognize equivalent conditions in Betula to have developed
gradually in the Populus borer. Those accepting this hypothesis are left with
the problem of explaining a sudden discontinuity (sympatric speciation) in the
flow of genes between the Populus borer and the newly-developing population
of the Betula-boring form. Explaining the irreversibility of the Populus-to-
Betula transition is also problematical.

It is much more logical to theorize that the switch was made by a form much
less specific to host conditions, and more specific to certain parts of a host
(e.g. certain sizes of limbs, or limbs versus trunk, etc.). Chemo-sensory
recognition of the host is not as crucial a matter in this case, allowing a little
more opportunity for evolutionary trial and error. Besides one has a possible
transitional form, or a fairly direct descendent of it, in Agrilus pensus.

Barter and Brown (1949) have suggested the pensus is allied with g. populi
Fisher, but their resemblance is merely superficial. The similarity of the
coloration in these two forms may reflect, in part, adaptation to somewhat
similar climatic conditions for their adult life spans, even though they occur
on opposite sides of the continent. The pensus aedeagus is unlike that of g.

1
Smith (1949a) suggested the reverse (i.e. that g. liragus was derived from anxius.

Carlson and Knight: Agrilus beetles 81

populi or g. liragus, contrary to Barter and Brown (1949). Although few g. pop-
uli males with the genetalia extruded were seen, we have examined the genetalia
of many g. liragus males. The aedeagus of pensus is actually more like that of

anxius than that of g- liragus (see figs. 5 & 8, above).

Agrilus pensus is related to anxius and des some subtle traits which also
link it with the Salix borers. The head is often flattened or broadly impressed
on the vertex as inq. niveiventris and the apices of the elytra often resemble
those of the Salix borers. About halfway between the midpoint and apices of
the elytra and near the suture, quadriguttatus and its subspecies often have
an elongated maculation formed by closely spaced pubescence (fig. 46c, above).
A similar maculation occurs in pensus (fig. 4c, above). The ecological! simi-
larities of alders and willows also render strong support to the hypothesis that
the transition from the Salicaceae to the Betulaceae was achieved by a species
from which pensus has descended rather directly. Figure 44 shows a Salix
boring species engendering an Alnus-Betula borer later than the evolution of
a Populus borer from a Salix borer. We postulate this because there are no
records suggesting that the Alnus-Betula borers are not endemic to the Nearc-
tic region.

The Populus Borers. There are, on the other hand, a number of Eurasian
forms in the Populus s borer group. One of them, A. ater L., may be more
closely allied to the Salix borers than any of the other Populus- boring forms.
Nevertheless, the listing of willow as a host for ater by Horion (1955) and
other taxonomists has dubious reliability. We have never seen a reference
which either cites a particular worker as having reared ater from willow or
gives other rearing records. One must be skeptical of these unverified records
for ater, because they may very well be referrable to guerini instead. The
biology of ater is discussed by Krogerus (1922) and Kangas (1942a, b); it ap-
pears to have habits roughly analogous to those of Nearctic Populus borers.

A. ater males do not have a median groove on the basal abdominal ster-
nites. In this respect ater appears to resemble guerini. In species which have
an abdominal groove, it appears to function in properly aligning males during
the act of probing and insertion of the aedeagus. In ater and guerini, the devel-
opment of very sharply pointed elytral apices the enlargement of the notch be-
tween them, may have functionally supplanted the abdominal groove of males.
The pointed elytra and absence of an abdominal groove in both species could
indicate a rather ancient relationship and a rather direct descent from the pair
of allied species proposed as the primordial Populus and Salix borers. How-
ever, the elytra of ater are shaped much like those of other Populus borers,
while those of guerini appear to be an elaboration of the usual Salix borer type.
Therefore, the extreme acuteness of the elytral tips and the associated loss
of the male abdominal groove may be convergent developments in these two
Species. It is unlikely that the shape of the elytral apices bears any functional
relationship to physical characteristics of the host. Hence, the consistent
difference in the elytra of Salix and Populus borers should accurately reflect
phyletic relationships.

Stepanov (1954) names poplars as hosts for A. fleischeri Obenberger, A.
ater juriceki Obenberger, and A. tamanukii Stepanov. He says that A. tschere-
panovi Stepanov lives on Populus s laurifolia, which is grouped with the balsam _
poplars. If A. fleischeri coreicus Kurosawa, A. fleischeri nipponicola Kuro-
sawa, and A. fleischeri kurosawai Obenberger have been properly classified
as subspecies of fleischeri, they are also Populus borers.

A. tamanukii probably relates more closely to Nearctic Populus borers

82 Contrib. Amer. Ent. Inst., vol. 4 no: 3, 1969

than other Eurasianforms. Like the Nearctic forms it has a median groove
on the basal abdominal sternites of males’. It was described from southern
- Sakkhalin, which, according to Tatewaki (1963), has a mixture of temperate
east Asiatic and subarctic phytogeographic elements. Therefore, this spec-
ies is an ecological counterpart of the Nearctic g. liragus and it is entirely
possible that tamanukii should also be ranked as a subspecies of granulatus.

Three Taiwanese species, which also have a male abdominal groove, are
put in the Populus borer group only on the basis of their descriptions. They
are A. subcuneiformis Kurosawa, A. satoi Kurosawa, and A. semiaurovitta-
tus Kurosawa. These species could be ecologic analogs or homologs of ce
granulatus and g. populi.

Obenberger (1958) believed that more aberration and subspeciation has oc-~-
curred among Paleartic buprestids than in those of the Nearctic region. Never-
theless, it seems unlikely that the number of Eurasian species and subspecies
of Populus borers could be as great as the number of available names. The
number of names needed for them will probably be reduced by synonomy in the
future.

Zoogeography

ASSUMPTIONS
Effect of Climatic Change. The climates to which each of these Agrilus

taxa are now adapted are probably similar to those in which they diverged
from their closest allies. Climatic change undoubtedly led to much of the geo-
graphic fragmentation precursory to their speciation and subspeciation. The
necessity of geographic isolation for speciation is generally acknowledged; mod-
els for sympatric speciation, except in a few special cases, appear unsupporta-
ble (Mayr, 1963).

Deevey (1949) says ‘‘the occurrence of closely related species in the same
area implies their differentiation in previous isolation; this isolation must have
lasted long enough, first to produce the reproductive divergence necessary to
avoid swamping of one by the other, and second to produce the ecologic differ-
entiation (differences in food habits, temperature preferenda, etc.) necessary
to avoid too stringent competition. Nearly all well studied cases of subspecia-
tion and speciation point to the Pleistocene as the time of previous isolation,
and the occurrence of closely related forms in the same area is therefore at-
tributed to post-Pleistocene alterations in geography and in biogeography. ’’

One must certainly expect exceptions to such a sweeping generalization as
this. However, the pitfalls to be encountered in its application are fewer than
are met with in trying to decide which pairs of species are really siblings and
which are not. Indeed, the important factors contributing to recent speciation
and subspeciation among Nearctic forms in the Agrilus anxius group seem to
have been the changing positions of glaciers, deserts, and prairies in relation
to forested areas and mountain ranges.

Pigmentation. Coloration, for the Agrilus in question, appears to be most-
ly a matter of climatic adaptation as opposed to being cryptic. The results of
studies among other groups of insects have interesting parallels which are

Neither Stepanov (1954) nor Kurosawa (1954) seem to have recognized this as a male
character, and may, consequently, have described males and females of the same
species under different names.

Carlson and Knight: Agrilus beetles 83

probably meaningful. Dobzhansky (1933) found that in certain Coccinellidae
there is a marked concentration of scarcely pigmented varieties in Turkestan,
Iran, California, and Mexico. The further a region is removed from these
centers, the darker is the average type of population found therein. The dark
est forms were found in eastern Siberia. Dobzhansky acknowledges a general
agreement of his findings with Allen’s rule, which according to him is more
properly called Gloger’s rule. This rule relates depigmentation of animals
to arid regions, and pigment accumulation to humid regions.

Netolitzky (1931) derived similar kinds of rules from studies of Carabidae.
He relates variations in pigmentation more to differences in altitude and lati-
tude than to humidity. According to his findings, pigmentation of European
carabid beetles increases with increasing altitude and latitude.

Kalmus (1941) gives some working hypotheses (he calls them laws) regard-
ing pigmentation that are more elaborate and combine some of the features of
Gloger’s rule with those of Netolitzky’s rules. He starts from a set of theoret-
ical premises which are based upon the structure of insect cuticle and the at-
tributes associated with the nature of its pigmentation. Kalmus says, ‘‘darken-
ing and hardening of the cuticle are one in the same process.’’ Whether or not
this is entirely true is still a controversial matter (Cottrell, 1964). Kalmus
(1941) attributes several different functions to hardening and darkening. He
maintains that the functional attributes of the cuticle which are enhanced with
increasing hardening and darkening are (1) heat absorption, (2) protection from
high frequency radiations (i.e. ultraviolet), and (3) moisture retention.

There seems to be no logical reason for contesting the validity of the as-
sumption that dark cuticles are more heat-absorbent than pale ones, and it
is supported, in a small way, by the experiments of Buxton (1924). Kalmus
(1941) generalizes that if the mode of life of an insect makes heat absorption
advantageous, its coloration will be dark; if heat absorption is irrelevant or
disadvantageous it may be pale. He also maintains that melanic forms are
found at high altitudes because their dark cuticles give them greater protection
from high frequency radiations. This may be true, but the effects of increas-
ing high frequency radiation and those of decreasing temperature with increas-
ing altitude cannot be mutually exclusive. His suggestion that insects exposed
to drought are dark seems to be antithetic to Gloger’s rule. Moreover, it might
be contradictory to relate hardening and darkening to aridity were it not for the
fact that iridescence is often associated with aridity. Many arid regions are
also hot, but iridescence would seem to be compensatory; according to Kalmus
(1941) it is associated with darkened cuticles, and gives them the heat absorb-
ing qualities of pale coloration in combination with the moisture retaining capa-
bilities accompanying heavy sclerotization.

Fisher (1928) was not cognizant of the close alliance between q. quadrigut-
tatus and q. niveiventris, but thought that each of these Salix borers was re-
lated to its sympatric Populus-boring counterpart, g. granulatus and g. populi,
respectively. The color resemblance of each Salix borer to the Populus borer
living with it is close enough so that they were usually lumped as single western
and eastern species prior to Fisher’s revision. While granulatus populi and
quadriguttatus niveiventris are iridescent reddish-cupreous above (the latter
sometimes very dark), the nominate subspecies are duller, being more or less
dusky olivaceous to reddish-fuscous. These pairs of Populus and Salix borers
seem to be the most austral and least alpine of the forms in the triad composed
of the Salix, Populus, and Alnus-Betula boring species.

The climatic affinities of anxius and g. liragus are quite boreal and alpine

84 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

by comparison. They are darkly pigmented, usually almost black, although
the pronotum and vertex of anxius are apt to be cupreous, particularly in spec-
imens from the northeastern U.S.A. and southeastern Canada. Fisher (1928)
failed to distinguish anxius from g. liragus because they are very similar.

The similarities within any of the sympatric pairs, anxius and granulatus
liragus, quadriguttatus niveiventris and granulatus populi, or nominate quadri-
guttatus and granulatus, are homoplastic', not homologous. They appear to be
due to adaptation to similar or equivalent climatic conditions. |

DISCUSSION

Faunal Connections. The gradual cooling of climate that occurred at the
close of the Tertiary, particularly at higher latitudes, brought southward re-
traction of the Arcto-Tertiary flora and largely eliminated it from Europe
(Spurr, 1964). Remnants of this forest persist in eastern Asia (Japano-Man-
churian), Pacific coastal North American (Vancouveran), and eastern North
America. North of these Arcto-Tertiary remnants, particularly in the more
continental regions, the modern boreal forest associations were formed (Lin-
sley, 1963). Logically, then, one looks to the Japano-Manchurian region for
faunal connections with the Vancouveran and Alleghenian regions. Linsley
(1963) shows that the cerambycid faunas of the latter two regions have separate
connections with the former. Species bridging the gap between browni and
horni or between the latter and criddlei may occur in the Japano- Manchurian
region. In this same region one might also find pre-Quaternary connections
with the non-boreal Salix and Populus borers of North America (cf p. 82).
These non-boreal forms may be rather direct descendents of the pre-Quater-
nary forms which presumably gave rise to the boreal ones, such as the Popu-
lus borers, tamanukii of eastern Asia, and g. liragus of North America. Each
of these boreal Populus borers could have had an independent origin but there
is no necessity for assuming that it was so.

Linsley (1963) says that Holarctic cerambycid species dominate in the
forests of such boreal derivatives of the Arcto-Tertiary Geoflora as the
northern and high elevation Pinaceae, and the northern deciduous trees, es-
pecially Salix and Populus. There is a definite likelihood that tamanukii is
conspecific with granulatus. In fact, tamanukii and g. liragus may represent
derivatives of a single boreal form which might have persisted at the southern
edge of the Bering Arc during late-Pleistocene glacial maxima.

Subspeciation. It is clear that g. liragus must have been isolated from g.
granulatus and g. populi (or the g. granulatus-g. populi ancestor) in an area of
colder climate, where it derived a more cold-adapted type of pigmentation and
its affinity for a boreal climate. There is no logical explanation for isolation
north of the g. granulatus-g. populi range unless it occurred as a result of
glaciation or unless g. liragus was absent from North America and derived its
boreal affinities in Asia, later migrating to North America through Beringia.
In the latter case, the problem of isolation from Japano- Manchurian forms is
raised.

According to Heusser (1965), macro-fossils, dated 35,000 B.P., from the
unglaciated Alaskan interior, show that a forest of aspen, birch, and spruce
persisted there throughout the Wisconsin glacial interval. This refugium seems
to be the only northern Nearctic one (other than Vancouver Island) for which

1
‘*Homoplasy is resemblance not due to inheritance from a common ancestry (Simpson,
1956).""

Carlson and Knight: Agrilus beetles 85

there is good evidence of the survival of forests during the Wisconsin glacial
maxima. This seems to be as likely a place as any for the derivation or re-
inforcement of the boreal affinities of g. liragus. As suggested above, another
possible refugium is the southern edge of the Bering-Chukchi platform which
was emergent during glacial maxima. Although most paleobotanists who have
studied the problem suggest that forests never existed on the Bering land
bridge, their views do not seem to be supported by the frequency of Holartic
distributions among boreal forest insects.

A. granulatus liragus occurs at the same latitudes with g. granulatus or
g. populi in mountainous regions, and in these areas the former seems to be
confined to higher altitudes, while g. granulatus or g. populi occur mostly in
river bottoms at lower altitudes. Their coloration appears to be a genuine in-
dication of their climatic affinities. Intergradation between g. liragus and
the other subspecies should occur at intermediate altitudes where cottonwood
(P. deltoides) or black cottonwood (P. balsamifera ssp. trichocarpa) occurs
together with aspen (P. tremuloides).!

One specimen from Crow Agency, Montana (July, 1916; R. K. Kellog),
recorded under Agrilus granulatus by Fisher (1928), appears to represent an
intergrade between g. granulatus and g. liragus, as its color is rather dark
and the spots on the elytra are not as distinct as cane of g. granulatus speci-
mens from farther south. The Agrilus granulatus ” discussed by Brown and
Stevenson (1963) are very similar to the foregoing. It is interesting that the
species of the various sections of the genus Populus hybridize in southern Al-
berta, where their habitats intergrade (Brayshaw, 1965). For g. granulatus
and g. liragus this area seems to represent a natural counterpart of the “‘hy-
bridized habitat’’ of Anderson (1948).

During the Pleistocene pluvials--the cool, moist periods when glaciation
occurred--the ranges of willows and cottonwoods must have expanded from
both east and west out into the southern Great Plains so that there was even
less disjunction than occurs presently. Although they do not show Salix and
Populus, the pollen analyses of Martin and Mehringer (1965) for the south-
western U. S., indicate that this is more than just a likelihood. Thus, the
stage would have been set for movement of eastern Salix and Populus borers
into the West and vice versa. Following continent-wide dispersal of these forms,
there may have been a sequence of hypsithermal divergences followed by broad
intergradations between eastern and western forms during pluvial periods. It
is very likely that this occurred several times, as Deevey (1949) says that in
the history of Lake Bonneville there have been two, perhaps three, major late-
Pleistocene pluvials.

At the present time the ranges of g. populi and g. granulatus do not appear
to interdigitate extensively. That of g. populi is western, extending from south-
ern California toBritish Columbia and as far east as Missoula, Montana. The
range of g. granulatus extends from the Great Plains eastward; although its
limits are not well-known, Fisher recorded it from as far north as Nassau,
New York, and as far west as Crow Agency, Montana, and south to Illinois and

wen Populus sp. or cultivar appears suitable for all of the subspecies of granulatus, pro-
vided that the former occur in habitats of the latter.

5 nena from these rearings are deposited in the Canadian National Collection, Ottawa,
Ontario.

86 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Missouri, and there is one specimen from North Carolina in the Horn collec-
tion’. In Colorado only g. granulatus is known, but some of the specimens?
examined from that area show indications that the two probably meet and inter-
grade there in places.

The Salix borers q. quadriguttatus and q. niveiventris are known from far-
ther south than their Populus-boring counterparts. The range of q. quadrigut-
tatus extends roughly from Georgia® to Texas, north to Colorado and east to
Pennsylvania and Washington, D. C. The northern range limit of q. niveiven-
tris lies just south of Placerville, California (see below) and, according to
Van Dyke (1942), this subspecies ranges south into Baja California: and east
into Arizona and New Mexico. Although we have seen no definite q. niveiven-
tris-q. quadriguttatus intergrades, some q. quadriguttatus specimens are dis-
tinctly tinged with red, probably indicating recent genetic exchange with q.
niveiventris. The divergence between these Salix borers was probably con-
current with that occurring between g. granulatus and g. populi.

Because of their color similarities, it appears that quadriguttatus niveiven-
tris and granulatus populi were confined to regions with similar climates during
their period of divergence (hypsithermal) from eastern allies. Their ranges
at that time may have been more nearly coextensive than at present. Although
q. niveiventris occurs in Baja California, Arizona, and New Mexico, g. populi
has never been collected in these areas, in spite of the occurrence of Free-
mont cottonwood (Populus freemontii) in their stream bottoms. On the
other hand, g. populi occurs undifferentiated well north of q. niveiventris,
where quadriguttatus is represented by the more northern subspecies, q. ful-
minans.

According to Van Dyke (1942) some records for q. niveiventris from British
Columbia may refer to q. fulminans, but this is not altogether certain because
g. populi was not distinguished from q. niveiventris any sooner than q. fulmin-
ans. We are not aware of any reliable records for q. fulminans outside of those
for Oregon and Northern California, and do not know how well-differentiated
q. niveiventris and q. fulminans are. Unfortunately, Fisher (1928) described
q. fulminans from a series of specimens which seem to represent intergrades*
between q. niveiventris and Salix borer populations north of the q. fulminans
type locality. This series (q. ful fulminans type and paratypes) was taken at Pia-
cerville, California, and Some specimens are iridescent reddish cupreous a-
bove, which seems fairly typical for q. niveiventris, while others are greenish.
Fisher (1928) had specimens from Biggs and Castle Crags in California which
were strikingly different from those collected farther south at Placerville. He

*Philadelphin Academy of Sciences, Philadelphia, Pennsylvania.

pe esonal collection S. G. Wellso, Dept. of Entomology, Michigan State University,
East Lansing.

*Record in Franklin and Lund (1956).

4 pisher based his distinction at q. fulminans upon the efflorescent spots on the sides of
the abdominal sternites, but this character does not seem to merit the value he has given
it and may be more characteristic of the population at the type locality than than of those
elsewhere.

| Carlson and Knight: Agrilus beetles 87

says, ‘‘the specimens from Biggs have the elytra dark green and the pronotum
more bronzy green, and two specimens from Castle Crags te ites , which I
have placed under this species, have the pronotum greenish blue and the elytra
a beautiful violaceous blue. ’’ Specimens were seen from Crescent
Mills, California, in the northeastern part of the Sierra Nevada, that are
bright apple-green. Barr’ said he had examined relatively few specimens of
this Pacific Northwest form, but those he had seen are green. Beer? said q.
fulminans specimens from northern California and Oregon are blue. A

Some sort of past range disjunction must account for the color differences
of q. niveiventris and q. fulminans. It may have occurred during a hypsither-
mal period, when the ancestral populations in the Cascades or northwestern
coast ranges may have been isolated from those in the southern coast ranges
or the Sierras. An alternative possibility is that the population that gave rise
to q. fulminans may have been isolated in a Pacific Northwest glacial refugium,
such as the one which, according to Heusser (1960), existed on Vancouver Is-
land. The nature of the climatic affinities of q. fulminans may become clearer
as more collections are made. :

The Multiplication of Species

Speciation in the anxius Group. The only cases of complete
speciation among the triad of the Populus, Salix, and Alnus-Betula borers that
can, with reliability, be designated as Nearctic, Pleistocene events are those
which led to the derivation of the ancestor of the latter and the subsequent form-
ation of two siblings from it. Previously (page 80) we discussed the possible
derivation of the ancestral form from a willow feeder. This may have occurred
very early in the Pleistocene, and the point of origin may have been eastern
North America. No species in the anxius group, other than its namesake, is
known to live on a species of Betulaceae in the western part of the continent.
However, Chamberlin (1924) says he collected a specimen of q. nhiveiventris
on Oregon alder (Alnus rubra) near La Grande, Oregon, and that a series was
collected on black cottonwood and Oregon alder at Oakridge, Oregon.

Following his description of g. populi, Fisher (1928) lists 4 of the paratypes
as having been taken on Oregon alder by Chamberlin at Oak Ridge, Oregon‘.
The only one of these paratypes remaining in the collection of the U. S. National
Museum has no host label. One in the Chamberlin collection at the California
Academy of Sciences, was labeled ‘‘Oakridge VI-29 Ore. /Populus trichocarpa/
W. J. Chamberlin Collector?.’’ Chamberlin (1929) described trichocarpae, a
synonym of g. populi from the same series of Oak Ridge specimens, but he did

1

Emendation of his ‘‘Castle Crag.’’

a

Personal communication, 1967; W. F. Barr, Dept. of Entomology, University of Idaho,
Moscow.

pansaual communication, 1967; F. M. Beer, Dept. of Biology, School of Science, Oregon
State University, Corvallis.

4
Emendation of Fisher (1928) who cites Oak Ridge as a California locality.

5
Personal communication, 1967; H. B. Leech, California Academy of Sciences, San Fran-
ciso.

88 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

not say any of these were collected on alder (cf above citation, Chamberlin,
1924). Beer and Hatch (1941) cite lombardy poplar, black cottonwood, and
Oregon alder as g. populi hosts, but are, no doubt, following Fisher (1928).

Beer! doubts the existence of a western form resembling g. populi that
infests Oregon alder. He has collected g. populi in places where alders and
black cottonwoods had been felled simultaneously, but saw none on the alders.
Moreover, he says, ‘‘Knowing Chamberlin’s inaccuracies on host plants--in
fact his general failure to mention this feature would prompt me to discount
completely occurrence on this plant [alder]. ”’

Evidently anxius has a greater specificity than pensus for hosts in a severe-
ly weakened condition, but anxius hosts are usually also larger and more likely
to occur on xeric sites. A. pensus, then, appears to be more ‘‘primary’’ in
its attacks on alders and young river birches than anxius is in attacking its
hosts. There is no 4 priori reason for assuming that anxius would not be at-
tracted to large felled or injured river birches, provided they were located
in the right places. A number of birches appear to be satisfactory for anxius,
and the alleged olfactory attractants, therefore, appear not to be so host-spec-
ies-specific as they are specific to certain conditions of the host. On the other
hand, there is no evidence to show that pensus is not attracted to such large
felled or injured river birches, but without such evidence one must assume that
it is not. It seems very necessary to hypothesize this and the other anxius-
pensus differences proposed above, if one is to explain their discreteness in
sympatry.

The distribution of Agrilus pensus suggests that it was confined to the At-
lantic seaboard during the primordial stage of its evolution. The divergence
of anxius and pensus may have occurred while the latter was confined to the
glacial refugium known to have existed in the southeastern U. S. There is
evidence showing that about 15 to 16 thousand years ago forests of birch and
alder mixed with pine, spruce and fir, existed in the Dismal Swamp region of
southeastern Virginia (Whitehead, 1965). Similar ecological conditions may
have existed there during glacial ages prior to the Wisconsin so that the com-
mon ancestor of anxius and pensus may have diverged from a Salix-boring
ancestor in the same refugium.

The ranges of pensus and anxius and also their color differences suggest
that when they were geographically isolated they were exposed to different
climatic conditions. Therefore, anxius may have been confined to a refugium
west of the Appalachians in the south central U. S. Although there has been
a great deal of dispute on this matter, it has become increasingly evident that
during the Wisconsin, at least, boreal tree species occurred as far south.as
Louisiana and Texas (cf Deevey, 1949 and Whitehead, 1965).

Following its continent-wide dispersal, anxius may have had its range
fractured during another glacial stage. Specimens from the East differ from
western ones in having the vertex and pronotum more distinctly cupreous-col-
ored. Specimens from the Upper Peninsula of Michigan may represent inter-
grades, as some are colored more like g. liragus, while the pronotum and
vertex of others are quite distinctly cupreous-tinged. Townes a says it is not

‘Deraannl communication; see footnote 3, page 87.

‘Daracnal communication, 1967; H. K. Townes, American Entomological Institute, Ann
Arbor, Michigan. ,

Carlson and Knight: Agrilus beetles 89

unusual for intergrades between eastern and western subspecies of ichneumon-
flies to occur in that area. These east-west differences of anxius adults could
relate to the fact that the only indigenous western birches of tree size are
those of the paper birch group, which are mostly boreal. Because anxius oc-
curs at least as far south as Washington, D. C. inthe East, adults from the
West, generally, may be more cold-adapted.

The Sibling Species Concept. Mayr (1963) defines sibling spec-
ies as ‘‘morphologically similar or identical populations that are reproduc-
tively isolated.’’ Certainly, anxius and g. liragus would be sibling species by
that definition, but it is of little value if it fails to make a distinction between
homoplasy ' and homology. ‘‘The fundamental difference between the method
of morphological and phylogenetic systematics is that the latter breaks up the
simple concept of resemblances (Hennig, 1965). ’’

The archaic level of Agrilus classification is due to the relatively unsophis-
ticated concepts of the classifiers, who usually have not tried to distinguish
between homoplasy and homology. Those unaware of this fact tend to regard
- Agrilus and other large buprestid genera as being more unique than they real-
ly are. Mayr (1963) says:

Among the 25 species of Carabus beetles from central Europe, 80 percent are poly-
typic, while in certain well known genera of buprestid beetles not a single” species is con-
sidered polytypic. There are still large groups of animals in which all species are listed
under binomials. It would be interesting to know to what extent this is due to lack of geo-
graphic variation in the phenotype and to what extent to insufficient taxonomic analysis.
Sibling species, of course, are nearly always monotypic.

Although Obenberger (1957) claimed that the Buprestidae are taxonomically
well-known, they certainly have not been studied nearly as intensively as the
majority of Carabidae. Therefore, Mayr’s comparison of buprestids and car-
abids is invalid. There appears to be no greater lack of geographical varia-
tion in the phenotypes of buprestids than in most other groups (see Obenberger,
1958). Binomial names are still preponderant over trinomials in Agrilus be-
cause much of the taxonomic work on that genus is old or has been done by
species describers instead of classifiers. Some of it has also heen done by
people working on local faunas. The forms they call ‘‘sibling species, of
course, are nearly always monotypic. ”’

Mayr (1963) drew the following ‘‘conclusions concerning sibling species’’:

1. There is no sharp division between ordinary species and sibling species...

2. Sibling species, when subjected to a thorough analysis, usually are shown to differ
in a whole series of minor morphological characters. Like ordinary species they are sep-
arated from each other by distinct gaps.

3. Sibling species are apparently particularly common in those kinds of animals in
which chemical senses (olfactory and so on) are more highly developed than the sense of

1
Defined in footnote on page 84.

2
Probably an overstatement of the case, with the possible exception of some smaller gen-
era. However, Mayr (1963) does not appear to be referring to any of the smaller genera
because they are obscure and because he elsewhere refers to Agrilus and Acmaeodera.

90 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

vision... Sibling species are apparently rarest in organisms such as birds that are most
dependent on vision in the recognition of epigamic characters.

4. There is no indication that sibling species arise by a process of speciation differ-
ent from that which gives rise to other species.

o. Degree of morphological similarity in sibling species is an indication not of genet-
ic Similarity, but rather of developmental homeostasis. A reconstruction of the genotype,
resulting in the reproductive isolation of two species, can take place without visible effect
on the morphology of the phenotype [cf number 2!].

6. Evolutionary changes in the genetic constitution seem to occur in groups of sibling
Species at the same rate as in groups of morphologically very distinct species.

The following paragraph from Mayr (1963) is certainly contradictory to num-
bers 4 and 6 of the conclusions listed above. He says:

In many groups of insects there are large genera with many species, each of which
appears to be limited to a single host. Genera with essentially monophagous species have
been described for microlepidoptera, solitary bees, buprestid beetles (Acmaeodera, Agri-
lus), chrysomelid beetles (Calligrapha, Arthrochlamys), and other groups. Two phenom-
ena in particular seem to suggest a mode of speciation in these food specialists that is
different from speciation in most other animals. The first is that monophagous insects
often, if not usually, belong to large genera. It is argued that sympatric speciation per-
mits more rapid and more frequent speciation than geographical speciation. It would seem
even more plausible, however, that the great number of congeneric species is due to the
vastly increased number of available niches and the reduction of competition. Likewise,
high food specificity should greatly enhance the efficiency of eco-geographical barriers.

Mayr seems to make analogies with speciation among insular birds, and in
doing so does not seem to be in complete accord with number 3 of his conclus-
ions about sibling species. He claims that:

As in all speciating animals the crucial events are apt to take place in peripherally
isolated populations and these have not been studied adequately in even a single one of
the groups of monophagous species. As a working hypothesis one might assume that a
subsidiary host may offer more favorable conditions under the changed ecological situa-
tion of the marginal environment (fig. [47]). The shift from one host to another will set
up an increased selection pressure that will result in rapid genetic alteration of the pop-
ulation. The amount of genetic variability appears to be low in a species that has been
adapted for life on a single host. Host specificity is thus an ideal prerequisite for rapid
Speciation.

A narrow-ranged host specificity certainly is low variability per se, but
does not necessarily reflect low genetic variability. There seems to be no rea-
son to assume that host specificity differs basically from other phenotypic as-
pects. Relatively invariable host specificities, therefore, may be a consequence
of homeostasis or coadaptation of the gene pool. ‘‘The real problem in specia-
tion is not how to produce a difference but rather how to escape from the cohe-
Sion of the gene complex. No one will comprehend how formidable a problem
this is who does not understand the power of the cohesive forces that are re-
sponsible for the coadapted harmony of the gene pool [Mayr, 1963].’’ As far
as we know, the problem of breaking away from gene pool coadaptation is great-
est for species with a narrow range of host specificity because the cohesiveness
of their gene pools may be greatest.

It appears that in the anxius group species which have highly epenaNaed re-
lationships to their hosts in a given respect (i.e. to species, size, physiologi-
cal condition, etc.) may usually evolve from less specialized species or from

Carlson and Knight: Agrilus beetles 91

Fig. 47. Primary hosts (capital letters) and subsidiary hosts (small letters)
of an essentially host-specific species. The large circle indicates the main range ©
of the species; the outlying small circles, the peripheral isolates. New primary
hosts (B, D) are acquired in some peripheral isolates, facilitating subsequent
speciation. Reinvasion from the isolate that is host specific for B will establish a
new species (hatched area) if reproductive isolation had been acquired during the
geographic isolation (legend as in Mayr, 1963; figure redrawn).

species specialized ina different way. The most recent (mid- to late-Pleisto-
cene) instances of complete speciation which seem to have occurred among the
Populus, Salix, and Alnus-Betula borer groups (fig. 44, above) are those in
which an Alnus-Betula boring species evolved from a Salix borer and subsequent-
ly gave rise to pensus and anxius. While anxius has evolved the greater speci-
ficity to physiological host condition, pensus appears to be more specific to

host size.

One does not need a special working hypothesis for explaining these spec-
lations. It is most plausible to attribute them to the events of the Pleistocene,
in particular glaciation, which undoubtedly gradually severed ancestral stocks
and left the disjunct populations with ranges that probably became separated
farther and farther and increasingly different in climate.

Mayr’s model (fig. 47) does not seem to provide for geographic separation
sufficiently great to result in the climatic differences that seem to have been
precursory to speciation in the anxius group, at least, nor does it seem to al-
low enough time for the occurrence of these climatic differences and for the
reconstruction of the genotype. Perhaps in groups for which rapid speciation
is a credible assumption, this model will prove to be worth considering. Most
of the large groups will require a good deal more study before it will be possi-
ble to assert that conventional hypotheses are not as plausible in specific cases

92 Contrib. Amer. Ent. Inst., Vol. 4, no. 3, 1969

as are hypotheses providing for rapid speciation.

Evolutionary Synopsis

_The reader is referred to figure 44 (p. 75 ) for a general summarization of
the relationships between the forms in the Agrilus anxius group. The impor-
tant points are these:

1) There appears to be no direct relationship between horni and anxius;
while horni is probably most directly allied to forms like browni, pensus is
the closest ally of anxius.

2) The major difference between horni and anxius (and allies of the latter)
relates to the distinctness of their modes of (hypothesized) host selection. A.
horni appears to rely upon a stem ‘‘measuring’’ mechanism in limiting the
placement of its eggs to aspen suckers of particular sizes (but apparently with-
out any discrimination in relation to their physiological condition). On the other
hand, anxius oviposits on hosts that are in a decrepit physiological state and is
apparently attracted to them by odiferous by-products of the bio-chemical break-
down associated with their decrepitude.

3) Forms linking these two distinct modes of life are found in pensus or
quadriguttatus (and its subspecies). These species are apparently less speci-
fic to host condition and more specific to host size than anxius, although prob-
ably more reliant upon a subnormal state of host physiology than horni.

4) A. quadriguttatus and guerini probably descended from a single phyletic
line, and the latter seemingly has developed some significant morphological
modifications from the usual Salix borer type. These Salix borers appear to
share common ancestry with: (a) criddlei, a gall-former living on willow,
which may in turn be somewhat related to horni, (b) a line of Populus borers
including the Palearctic ater and Nearctic granulatus, and (c) anxius and pen-
sus, through a later derivation independent of that for the Populus borer group.

5). The sibling species concept has been applied far too indiscriminately
in advance of a thorough assessment of the relationships among species, and
this is certainly pertinent to many groups besides Agrilus. The following pairs
have often been related by homoplasy (similarity) rather than homology: anxius

and g. liragus; g. granulatus and q. quadriguttatus; or granulatus populi and

quadriguttatus niveiventris.
6) The only recent speciations that seem evident for the Populus, Salix

and Alnus-Betula borer groups are the derivation and subsequent splitting of
an anxius-~pensus common ancestor.

7) No special models are needed to explain these speciations or the sub-
speciations that have occurred in quadriguttatus (Nearctic Salix borers) or
granulatus (Nearctic Populus borers). They can logically be related to biogeo-
graphical events of the Pleistocene.

Carlson and Knight: Agrilus beetles 93

SUMMARY

The principal objective of this rai was to assess the relationship between
A. anxius, A. pensus, A. granulatus liragus, and A. horni. This was accom-
plished by: (1) studying specimens, (2) assembling y what was known about gen-
eral aspects of their biology, augmenting it in places with new observations,

(3) making behavioral observations on the adults in the field and laboratory,
and (4) analyzing their phylogeny and zoogeography, relying cpinen | on mater-
ial from the preceeding sections.

In the process of studying the taxonomy of the adults, the re for a number
of nomenclatural changes became apparent, Changes which were made very re-
cently (see Carlson, 1969) include lowering four taxa from species to subspecies
status (populi and liragus to ssp. of granulatus, niveiventris and fulminans to
ssp. of quadriguttatus) and the suppression of two names, betulae and nevaden-
sis as synonyms of pensus and q. niveiventris, respectively. Herein lateralis
is suppressed as a synonym of ruficollis, and the species which has been incor-
rectly called lateralis is named browni.

Barter and Brown (1949) distinguished g. liragus from anxius, but thought
horni to be a probable synonym of the latter. After Harrison (1959) had dis-
covered horni boring in the roots of aspen suckers, Nord et al. (1965) estab-
lished its identity and showed that horni was clearly distinct from either anxius
or g. liragus.

A careful comparison of the adults revealed that horni has some previously
unnoticed or unemphasized morphological traits which, although they are sub-
tle, set it apart from anxius or g. liragus. They are discussed at some length
in the taxonomy section and again in the evolution section, where their signifi-
cance is assessed. |

These anomalous morphological characteristics of horni seemed to link it
in some way with species which lack prehumeral carinae. A paper by Fisher
(1922), completely overlooked until late in this study revealed that the habits
of A. browni are remarkably like those of horni. The morphological traits
separating horni and anxius or g. liragus and linking it with other forms, not
closely related to the latter, thus became more meaningful. The important
ways in which the habits of horni differ from those of anxius or g. liragus are
enumerated below. i

1) The eggs of horni are deposited at the bases of young aspen suckers,
while suitable oviposition sites for g. liragus and anxius occur almost any-
where along the stems of larger trees.

2) After hatching the horni larvae bore into the phloem and proceed down-
ward to a root and move rather directly out along it for a distance of up to 3
1/2 feet before the first molt occurs. This primary molt probably does not
occur until the spring following eclosion from the egg, when the larvae may
begin spiraling back toward the sucker stem. The first molt of anxius or g.
liragus larvae occurs soon after the larvae bore into the phloem. Mi

3) In many cases the larvae of anxius and g. liragus complete their devel-
opment and form a pupal chamber during the same year in which eggs are laid.
They thus have a one-year life cycle under some conditions and a two-year
cycle under others. The usual length of life cycle is two years for horni, and
the only exception that seems possible, is an occasional three-year cycle.

4) While anxius and g. liragus infest hosts that are under physiological
stress and seem to be attracted to them by odiferous products of biodegreda-
tion, horni appears to distinguish suitable from unsuitable hosts on the basis
of size rather than their physiological condition. The suckers on which horni

94 Contrib. Amer. Ent. Inst., Vol. 4, no. 3, 1969

eggs are found seem to be of normal vigor (i.e., no different than their neigh-
bors). Sucker diameter (as opposed to height) appears to be the most impor-
tant variable correlating with the occurrence of eggs. It is, therefore, sug-
gested that horni restricts its egg deposition to suckers within rather narrow
diameter limits (0.3 to 0.9 inches at base) by thigmotactic means. The dif-
ferences between the two kinds of host selection seem to be so extensive as to
preclude the derivation of one from the other except through a series of inter-
mediate stages.

Agrilus pensus seems to be a direct descendant of an ancestor representing
one of the intermediate stages. A two-year life cycle seems prevalent, and
a one-year cycle is unlikely because of the small size and thin bark of its hosts
(alders and ae river birches). A. pensus is probably more primary than
anxius or g. liragus, although it seems likely that the hosts it attacks are un-
der some physiological stress. Nevertheless, the average level of host resis-
tance which pensus Ss is subjected to should approximate the maximum encountered
by anxius or g. liragus.

“Host resistance is treated here as a population regulating factor, together
with parasites and predators. The sinuate galleries of anxius and zigzag gal-
leries of g. liragus appear to be mechanisms for coping with host resistance, ©
because the galleries of either species follow a more random course in the
phloem of hosts that have succumbed or are near death.

The survival of horni larvae may depend, in part, upon the ability to lower
the vigor of the host through root-girdling. Their ability to significantly lower
the vigor of the host by root-girdling should relate to root diameter. Because
root diameter seems to correlate well with sucker diameter, the restriction of
Oviposition by horni to suckers with basal diameters ordinarily below 0.9 inches
is probably an indirect means for coping with host resistance.

The effects of parasites and predators as population regulating factors is
more obvious, and for this reason, their significance may be overrated in
comparison with that of host resistance. Woodpeckers and egg-parasitizing
chalcidoids account for the greatest percentage of combined parasite and pred-
ator effects upon populations of anxius and g. liragus. The eggs of horni are
parasitized rather infrequently, but larval parasites consume a very high per-
centage (about 42 percent) of horni brood between the late larval and late pupal
stages. Some corrections are made regarding incorrect parasite rearing rec-
ords.

Even though horni and g. liragus live on the same host species here appears
to be no chance for interspecific mating because the males of each orient to
hosts that are likely to be attractive to conspecific females. Moreover, horni
females are completely unreceptive to liragus males, at least in the labora-
tory, and this seems to partially preclude hybridization in chance encounters.
It appears, that g. liragus males have lost some subtle facet from their court-
‘fn behavior repertoire. A. anxius females are occasionally receptive to g.

liragus males, but horni females, nevertheless, seem quite receptive to anxius
~males. This seems to put the male- -to-female communications of anxius some-

where intermediate between those of g. liragus and horni, as horni males are

accepted readily by females of either anxius or g. liragus. A. anxius could

be intermediate in this respect because of the relative propinquity of its ances-

try to species that live on willows as compared to that of g. liragus.

Males show no disinclination to court females of other species in the lab-
oratory. Moreover, males do not even appear to recognize the sex of the in-
dividuals they court until they have made physical contact with them. Sexual

Carlson and Knight: Agrilus beetles 95

recognition by males appears to depend upon the structure of the courted in-
dividual’s last abdominal segment. That segment is sexually dimorphic, and
after a courting male probes it with his aedeagus he leaves males but prolongs
courtship even with unreceptive females. This seemingly inefficient mode of
sexual recognition appears to be homostatically tied to a dispersing function
for homosexual encounters between the quasi-territorial males.

The interpretation presented here concerning evolutionary relationships
within the Agrilus anxius group, contrasts with views held by earlier students.
For example, horni is not considered as being a close ally of anxius or gran-
ulatus liragus. A. horni and browni have very similar habits, and these
Species appear to be more closely related to each other than to other species.
This interpretation leaves a distinct phyletic gap between horni and anxius or
g. liragus. Intermediates which seem to fill this gap are criddlei and a group
of willow boring species that includes quadriguttatus. The ancestors of the
latter group seem to have given rise first to a phyletic line that includes g.
liragus and later to one including both anxius and pensus. The derivation of
the line including anxius and pensus, and its subsequent splitting seem to be
the only recent (mid- to late-Pleistocene) cases of complete speciation for
the part of the anxius group considered in detail.

ACKNOWLEDGMENTS

This study was supported by the U. S. Department of Agriculture, McIntire-
Stennis Forestry Research Fund. We thank Mr. George B. Vogt of the Agri-
culture Research Service, Entomology Research Division, for his kind assist-
ance on a number of matters, including the borrowing of specimens from the
U. S. National Museum and Museum of Comparative Zoology. We also thank
Dr. John F. Lawrence of the latter institution for his cooperation. Persons
whose help on various matters is unacknowledged in the text, but nonetheless
much appreciated, are: Mr. Hideki Hagiwara, Mr. Josef N. Knull, Dr. Rich-
ard R. Mason, Dr. Thomas E. Moore, Dr. John C. Nord, Dr. Daniel Otte,
Dr. Henry K. Townes, Dr. Marjorie C. Townes, and Dr. H. R. Wong. Mrs.
Ann Wolff typed the manuscript for photo-offset printing.

96 Contrib. Amer. Ent. Inst., Vol. 4, no. 3, 1969

LITERATURE CITED

Alexander, R. D., T. E. Moore, and R. E. Woodruff 1963. The evolutionary
differentiation of stridulatory signals in beetles (Insecta: Coleoptera).
Anim. Behav. 11: 111-115.

Alexeev, A. V. 1960. On the morphology and systematics of larvae of some
species of Agrilus Curt. in the European part of U. S. R. R. Zool. Zhur.
39: 1497-1510, Eng. Summ.

Anderson, E. 1948. Hybridization of the habitat. Evolution 2: 1-9.

Anderson, R. F. 1944. The relation between host condition and attacks by the
bronzed birch borer. J. Econ. Entomol. 37: 588-596.

Arru, G. M. 1962. The influence of soil conditions on the vulnerability of the
poplar to Agrilus viridis L. var. populnea Schaef. (Coleoptera: Bupresti-
dae) in northern Italy. World For. Congr. Proc. 5: 948-950, Fr., Eng.
Summ.

Balch, R. E. and J. S. Prebble 1940. The bronze birch borer and its rela-
tion to the dying of birch in New Brunswick forests. For. Chron. 16: 179-
201.

Barter, G. W. 1957. Studies of the bronze birch borer, Agrilus anxius Gory,
in New Brunswick. Canad. Entomol. 89: 12-36.

1965. Survival and development of the bronze poplar borer,
Agrilus liragus Barter & Brown (Coleoptera : Buprestidae). Canad. Ento-
mol. 97: 1063-1068.

and W. J. Brown 1949. On the identity of Agrilus anxius Gory
and some allied species (Coleoptera : Buprestidae). Canad. Entomol. 81:
245-249.

and D. G. Cameron 1955. Some effects of defoliation by the
forest tent caterpillar. Canada Dept. Agr., For. Biol. Div., Bi-Monthly
Prog. Rept. 11 (6): 1.

Basurmanova, O. K. 1958. Biological forms of Agrilus viridis L. Zool. Zhur.
37: 1039-1044, Eng. Summ.

Beer, F. M. and M. H. Hatch 1941. The Coleoptera of Washington: Bupresti-
dae. Univ. Wash. Pub. in Biol. 10: 93-142.

Benoit, P. 1965. Morphologie larvaire des Agrilus liragus Barter et Brown
et Agrilus anxius Gory (Coleoptera: Buprestidae). Canad. Entomol. 97:
768-773.

Bier, J. E. 1964. The relation of some bark factors to canker susceptibility.
Phytopath. 54: 250-253.

Blanchard, F. 1889. A list of the Buprestidae of New England. Entomol. A-
mer. 5: 32.

Brayshaw, T. C. 1965. Native poplars of southern Alberta and their hybrids.
Canad. Dept. For. Pub. no. 1109, 40 p.

Britton, W. E. 1922. The bronze birch borer (22nd Rept. of the State Ento-
mologist for 1922). Conn. Agr. Exp. Sta. Bul. 247: 359-361.

Brooks, F. E. 1914. The apple root borer. J. Agr. Res. 3: 179-185.

Carlson and Knight: Agrilus beetles 97

Brown, C. E. and R. E. Stevenson 1963. Forest insect conditions in Alberta.
Ann. Rept. For. Ins. and Dis. Surv., For. Entomol. and Pathol. Br.,
Canada Dept. For., for 1963, p. 100.

Burks, B. D. 1967. In: Hymenoptera of American north of Mexico, synoptic
catalog. U. S. Dept. Agr., Agr. Monogr. no. 2, suppl. 2, K. V. Krom-
bein and B. D. Burks, eds., p. 268.

Burril, T. J. 1890. The lombardy poplar borer, Agrilus granulatus Say. In:
5th Rept. U. S. Entomol. Comm. (House Misc. Doc., vol. 41), A. S. Pack-
ard, ed. pp. 443-444.

Buxton, P. A. 1924. Heat, moisture, and animal life in deserts. Proc. Royal
Soc. Lond. (ser. B) 96: 123-131.

Carlson, R. W. 1968. Biology, taxonomy, and evolution of four sympatric
Agrilus beetles (Coleoptera: Buprestidae). Ph. D. dissertation, Univ. of
Michigan, Ann Arbor, 232 p.

1969. Title as in Carlson (1968). Diss. Abstr. 29: 2930B.

Chamberlin, W. J. 1924. Notes on the Buprestidae of Oregon with descriptions
of n. sp. J. N. Y. Entomol. Soc. 32: 185-194.

1929. Remarks on Buprestidae (Coleoptera) of the North
Pacific Coast Region with descriptions of new species. Pan-Pac. Entomol.
): 409-116,

Chittendon, F. H. 1898. A destructive borer enemy of birch trees, with notes
on related species. U. S. Dept. Agr., Div. Entomol. Bul. (n.s.) 18: 44-51.

1900. Food plants and injury of North American species of
Agrilus. U. S. Dept. Agr., Div. Entomol. Bul. (n.s.) 22: 64-68.

Cook, A. J. 1890. The willow buprestid (Agrilus torpidus). Rept. Mich. State
Bd. Agr. 29: 102-128.

Cottrell, C. B. 1964. Insect ecdysis. In: recent advances in insect physiolo-
ey, vol. 2, J. W. L. Beament, J. E. Treherne, and V. B. Wigglesworth,
eds. Academic Press, London, op. 479-218.

Davis, D. W. and N. N. Raghuvir 1964. The biology of the rose stem girdler,
Agrilus rubicola communis, on raspberries in Utah (Coleoptera : Bupresti-
dae). Ann. Entomol. Soc. Amer. 57: 154-159.

Davis, G. C. 1891. Notes ona few borers. Insect Life 4: 64-67.

Davis, G. C. 1892. Notes onafew borers. 22nd Ann. Rept. Entomol. Soc.
Ont., pp. 80-82. ;

Deevey, E. S. 1949. Biogeography of the Pleistocene. Bul. Geol. Soc. Amer.
~ 60: 1315-1416.

Dejean, Comte P. F. M. A. 1833. Catalogue des Coléoptéres de la collection
de la collection de M. le comte Dejean, pt. 1. Méquigon-Marvis, Paris,
96 p. :

1836. Catalogue des Coléoptéres de la collection
de M. le comte Dejean, 3rd edit., pt. 1-4. Méquignon-Marvis, Paris, 384 p.

Dobzhansky, Th. 1933. Geographical variation in lady-beetles. Amer. Nat.
67: 97-126.

1956. What is an adaptive trait? Amer. Nat. 90: 337-347.

98 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Engel, R. L. 1968. Some comparative observations on the biology and habits
of four species of the genus Agrilus (Coleoptera : Buprestidae). Unpublished
M. F. Thesis, Univ. of Michigan, Ann Arbor, 38 p.

Esau, K. 1953. Plant anatomy. John Wiley and Sons, Inc., New York, 1735 p.
Fabricius, J. C. 1787. Mantissa Insectorum I: 184-185.

Fisher, W. S. 1922. Notes on Agrilus lateralis Say (Coleop.). Proc. Ento-
mol. Soc. Wash. 24: 124-125.

1928. A revision of the North American species of buprestid
beetles belonging to the genus Agrilus. U. S. Nat. Mus. Bul. 145, 347 p.

Fowler, R. F. 1963. Vigor and density of aspen suckers and attack by Agrilus
n. sp. Unpublished M. F. thesis, Univ. of Mich., Ann Arbor, 42 p.

Franklin, R. T. and H. O. Lund 1956. The Buprestidae (Coleoptera) of Geor-
gia. Ga. Agr. Exp. Sta. Tech. Bul. (n.s.) 3, 48 p. :

Frost, C. A. 1912. New species of Coleoptera of the genus Agrilus. Canad.
Entomol. 44: 250-252.

Gaudin, L. 1921. Captures de Coléoptrés de France. Bul. Soc. Entomol. Fr.
1921: 114-116. |

Gory, H. 1841. Supplement Buprestidae. Vol. 4 of: Histoire et Incongraphie
des Insects Coleopteres, by F. L. Laporte de Castelnau and H. Gory. P.
Duménil, Paris, 356 p.

Graham, K. 1959. Release by flight exercise of a chemotropic response from
photopositive domination in a scolytid beetle. Nature 184: 283-284.

1961. Air swallowing: a mechanism in photo reversal of the bee-
tle Trypodendron lineatus. Nature 191: 519-520.

Graham, S. A., R. P. Harrison, and C. E. Westell, Jr. 1963. Aspens, phoe-
nix trees of the Great Lake Region. Univ. of Mich. Press, Ann Arbor,
272 p.

Hall, R. C. 1933. Post-logging decadence in northern hardwoods. Univ. of
Michigan Sch. For. and Cons. Bul. 3, 66 p.

Harlow, W. M. and E. S. Harrar 1958. Textbook of dendrology, 4th ed. Mc-
Graw-Hill Book Co. ; Inc.,:N. Y.).561 p.

Harris, T. W. 1829. Contributions to entomology VI. New England Farmer
8 (1): 2-3. :

Harrison, R. P., Jr. 1959. Insects and diseases of aspen. Ph. D. disserta-
tion, Univ. of Michigan, Ann Arbor, 259 p.

Heering, H. 1956. Zur Biologie, Okologie und zum Massenwechsel des Buch-
enprachtkafers (Agrilus viridis L.). Ztsch. Angew. Entomol. 38: 250-287;
39: 76-114.

Hennig, W. 1965. Phylogenetic systematics. Ann. Rev. Entomol. 10: 97-116.

Heusser, C. J. 1960. Late-Pleistocene environments of North Pacific North
America. Amer. Geogr. Soc. Spec. Pub. no. 35, 308 p.

1965. A Pleistocene phytogeographical sketch of the Pacific
Northwest and Alaska. In: The Quarternary of the United States, H. E.
Wright and D. G. Frey, eds. Princeton Univ. Press, Princeton, N. J.,
pp. 469-483.

Carlson and Knight: Agrilus beetles 99

Hitchcock, C. L., A. Cronquist, M. Owenby, and J. W. Thompson 1964. Vas-
cular plants of the Pacific Northwest, part 2. Univ. of Wash. Press, Se-
attle. 597 p.

Horion, A. 1950. Diskontinuerliche Ost-West Verbreitung mitteleuropaichen
Kafer. Proc. 8th Int. Congr. Entomol., Stockholm, pp. 408-417.

1955. Faunistik der mitteleuropaischen Kafer, vol. 4 Kommis-
sionverlag, Erwald Reitter, Munich, 280 p.

Horn, G. H. 1891. The species of Agrilus of boreal America. Trans. Amer.
Entomol. Soc. 18: 277-336.

Hutchings, C. B. 1923a. Some biological observations on the bronze birch
borer, Agrilus anxius Gory. 15th Ann. Rept. Quebec Soc. Prot. Plants
(1922-1923), pp. 89-92.

1923b. Some notes on the biology of two buprestids infest-
ing blackberry and hazel. 53rd Ann. Rept. Entomol. Soc. Ont., pp. 43-46.

Kalmus, H. 1941. Physiology and ecology of cuticular color in insects. Na-
ture (London) 148: 428-431.

Kangas, E. 1942a. Agrilus ater L., als Espenschadling in Finland. Ztsch.
Angew. Entomol. 28: 359-365.

Kangas, E. 1942b. Forstentomologische Studien an der Espe. Ann. Entomol.
Fennici 8: 49-71.

Kennedy, J. S. and C. O. Booth 1963. Coordination of successive activities
in an aphid. The effect of flight on the settling responses. J. Exp. Biol.
40: 351-369.

Kerremans, C. 1900. Buprestides noveaux et remarques synonomiques. Ann.
Soc. Entomol. Belg. 44: 282-351.

Kessler, G. 1958. Zur Characterisierung der Siebrohrenkallose. Ber. Schweiz.
Bot. Ges. 68: 5-43.

Knight, F. B. 1958. The effects of woodpeckers on populations of the Engel-
mann spruce beetle. J. Econ. Entomol. 51: 603-607.

and W. Albertin 1966. Portable X-ray equipment: a useful tool
in forest entomological field research. Pap. Mich. Acad. Sci., Arts, and
Let. S24 pty: 1)0°28-26)

Knull, J. N. 1922. Annotated list of the Buprestidae of Pennsylvania (Coleop-
tera). Canad. Entomol. 54: 79-86.

1930. Notes on Coleoptera, No. 2. Entomol. News 41: 82-86.
1934. Notes on Coleoptera, No. 4. Entomol. News 45: 207-212.

1950. In: Insect enemies of eastern forests, F. C. Craighead,
ed. U. S. Dept. Agr. Misc. Pub. 657: 187-188.

Kramer, P. J. and T. T. Kozlowski 1960. Physiology of trees. Mc-Graw-
Hill Book Co., Inc., New York, 642 p.

Kozlowski, T. T. 1964. Water metabolism in plants. Harper and Row, Pub. :
New York, 227 p.

100 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Krogerus, R. 1922. Studien uber Agrilus-Arten, I und II. Not. Entomol.
(Helsinki) 2: 10-14, 109-113.

Kurosawa, Y. 1954. Buprestid fauna of eastern Asia (2). Tokyo Nat. Sci. Mus.
Bul. (n.s.) 1: 82-93.

1956. Buprestid fauna of eastern Asia (3). Tokyo Nat. Sci.
Mus. Bul. (n.s.) 3: 33-41.

Larsen, J. 1901. A disease of the white birch. 3rd Rept. Michigan Acad.
Sci., Arts, and Let., pp. 46-49.

Lecomte, C. 1925. Sur quelques Coleoptéres obtenus d’élevages. Bul. Soc.
Entomol. France, pp. 294-296.

Leconte, J. L. 1859. Revision of the Buprestidae of the United States. Trans.
Amer. Phil. Soc. (2nd ser.) 11: 187-258.

Lindroth, C. H. and E. Palmén 1956. In: Taxonomist’s glossary of genetalia
in insects, S. L. P. Tuxen, ed. E. Munksgaard, Copenhagen, pp. 69-76.

Linsley, G. E. 1963. Bering Arc relationships of Cerambycidae and their
host plants. In: Pacific Basin biogeography, a symposium of the Tenth
Pac. Sci. Congr. (1961), J. S. Gressit, ed. Bishop Museum Press, Hono-
lulu, pp. 159-178. |

Lintner, A. J. 1889. Fifth report on injurious and other insects of New York.
42nd Rept. N. Y. State Mus., p. 317.

Lundberg, S. 1962. Contribution to the knowledge of Swedish Coleoptera 5,
Agrilus guerini Lac. Entomol. Tidsikr. 83 (3/4): 172-174, Sw.

Manee, A. H. 1913. Observations on Buprestidae at Southern Pines, North
Carolina. Entomol. News 24: 167-171.

Martin, P. S. and P. J. Mehringer, Jr. 1965. Pleistocene pollen analysis
and biogeography of the Southwest. In: The Quarternary of the United
States, H. E. Wright and D. G. Frey, eds. Princeton Univ. Press, Prince-
ton, N. J., pp. 433-451.

Mayr, E. 1963. Animal species and evolution. Belknap Press, Harvard Univ. ,
Cambridge, Mass., 797 p.

Megerle von Muhlfeld, J. K. 1823. In: Coleoptera und Lepidoptera, ein system-
atishes Verzeichness mit beigesetzten Preissen der Vorrathe, G. Dahl, ed.
Vienna, Akkermann, pp. 19-20.

Mundinger, F. G. 1941. Two buprestid cane-borers of brambles with experi-
ments on control. J. Econ. Entomol. 34: 532-537.

Nash, R. W., E. J. Duda, and N. H. Gray 1951. Studies on extensive dying,
regeneration, and management of birch. Maine For. Serv. Bul. 15, 82 p.

Netolitzky, F. 1831. Einige Regeln in der geographischen Verbreitung geflu-
gelten Kaferrassen. Biol. Zentralbl. 51: 277-290.

Nord, J. C. 1962. A biological study of a new species of Agrilus (Coleoptera:
Buprestidae) Unpublished M. F. thesis, Univ. of Michigan, Ann Arbor,
61 p.

Carlson and Knight: Agrilus beetles 101

Nord, J. C., F. B. Knight, and G. B. Vogt 1965. Identity and biology of an
aspen root girdler, Agrilus horni. For. Sci. 11: 33-41.

Obenberger, J. 1957. Quelques considérations sur les sous-genre Epinagri-
lus v. Stepanov et sur les caractéres du genre Agrilus Curtis en général
(Col. Bupr.). Sb. Entomol. Oddel. Narod. Mus. v Praz (Acta. Entomol.
Mus. Nat. Pragae) 31: 77-89.

1958. Sur les relations entre les Buprestides de la région
paléarctique et néarctique. Proc. Tenth Intern. Congr. Entomol. (1956)
1: 213-216,

1959. Sur les espéces du genre Agrilus Curtis de 1’ Australie
et Océanie (Coleoptera, Buprestidae). Sb. Entomol. Oddel. Narod. Mus.
v Praz (Acta. Entomol. Mus. Nat. Pragae) 33: 233-240.

Ohgushi, R. 1963. Development of ovary and pre-oviposition period of citrus
flat-headed borer, Agrilus auriventris auriventris S. Saunders under labor-
atory conditions. Jap. J. Appl. Entomol. Zool. 7: 92-96, Jap., Eng. summ.

Saunders, E. 1871. Catalogus Buprestidarium, p. 126.

Say, T. 1823. Descriptions of coleopterous insects collected in the late expedi-
tion to the Rocky Mountains. J. Acad. Nat. Sci. Philadelphia 3: 139-216.

Schaefer, L. 1949. Buprestides de France. Misc. Entomol., suppl., 511 p.

Semenov~-Tian-Shanskij, A. 1935. Analecta coleopterologica. Entomol. Oboz-
renie (Rev. d’entomol. U. R. S. S.) 35: 276, Latin.

Siegel, S. 1956. Nonparametric statistics for the behavioral sciences. Mc-
Graw-Hill Book Co., Inc., New York, 312 p.

Smith, J. B. 1892. Notes on blackberry borers and gall-makers. Insect Life
a 2130,

Smith, 8S. G. 1949a. Evolutionary changes in the sex chromosomes of Coleop-
tera. Nature (London) 164: 237-238.

1949b. Evolutionary changes in the sex chromosomes of Coleop-
tera I. Wood borers of the genus Agrilus. Evolution 3: 344-357.

Spaulding, P. and H. J. MacAloney 1931. A study of organic factors concerned
in the decadence of birch on cut-over lands in northern New England. J.
Forestry 29: 1134-1149.

Spurr, S. H. 1964. Forest ecology. Ronald Press, New York, 352 p.

Stager, R. 1923. Ein Schadling der Himbeerkulturen. Schweiz. Entomol. Anz.
a. 21-22, 25-26, 29730,:33-35, |
Stepanov, V. N. 1954. Two species of the genus Agrilus Curt. subgenus Epin-

agrilus sub-gen. nov., new to the fauna of the USSR. Zool. Zhur. 114-119.

Swaine, J. M. 1918. A new forest insect enemy of white birch. Canad. J.
For. 14: 1928-1929.

Tanner, V. M. 1927. A preliminary study of the genetalia of female Coleop-
tera. Trans. Amer. Entomol. Soc. 53: 5-50.

Tatewaki, M. 1963. Phytogeography of the islands of the North Pacific Ocean.
In: Pacific Basin biogeography, a symposium of the Tenth Pac. Sci. Congr.
(1961), J. L. Gressit, ed. Bishop Museum Press, Honolulu, pp. 23-28.

102 Contrib. Amer. Ent. Inst., vol. 4, no. 3, 1969

Théry, A. 1942. Faune de France Volume 41, Coleoptérés Buprestides. Paul
Lechevalier, Paris, 221 p.

- Townes, H. K. Jr. 1944. A catalogue and reclassification of Nearctic Ichneu-
monidae. Mem. Amer. Entomol. Soc. 11, pt. 1, 477 p.

Townes, H. K. and M. Townes 1951. In: Hymenoptera of America north of
Mexico, a synoptic catalog. C. F. W. Muesebeck, K. V. Krombein and
H. K. Townes, eds. U. S. Dept. Agr., Agr. Monogr. no. 2, pp. 184-410.

1960. Ichneumon-flies of America north of
Mexico 2. Subfamilies Ephialtinae, Xoridinae, and Acaenitinae U. S. Nat.
Mus. Bul. 216, pt. 2, 676 p.

1962. Ichneumon-flies of America north of
Mexico 3. Subfamily Gelinae, tribe Mesostenini. U. S. Nat. Mus. Bul.
216, pt. 3, 602 p.

U. S. Weather Bureau 1964. Climatological data for Michigan (1964).

VanDyke, E. C. 1942. Contributions toward a knowledge of the insect fauna
of Lower California (Coleoptera: Buprestidae). Proc. Cal. Acad. Sci.
24: 97-132.

vité, J. P. 1961. The influence of water supply on oleoresin exudation pres-
sure and resistance to bark beetle attack in Pinus ponderosa. Contr.
Boyce Thompson Inst. 21: 37-66.

Walton, R. R. 1951. Control of the red-necked cane borer on prostrate bram-
bles by summer pruning. J. Econ. Entomol. 44: 950-954.

Whitehead, D. R. 1965. Palynology and Pleistocene phytogeography of un-
glaciated eastern North America. In: the Quarternary of the United States,
H. E. Wright and D. G. Frey, eds. Princeton Univ. Press, Princeton,
N. J... pp. 411-4382;

Wigglesworth, V. B. 1964. The hormonal regulation of growth and reproduc-
tion in insects. In: recent advances in insect physiology, vol. 2, J. W.
L. Beament, J. E. Treherne, and V. B. Wigglesworth, eds. Academic
Press, London, pp. 247-336.

Wong, H. R. and B. B. McLeod 1965. The host and distribution of Agrilus
criddlei Frost in Canada. Canada Dept. For., For. Entomol. and Pathol.
Br., Bi-Monthly Prog. Rept. 21 (4): 3.

Zahner, R. and N. V. DeByle 1965. Effect of pruning the parent root on
growth of aspen suckers. Ecology 46: 373-375.

and J. R. Donnelly 1967. Refining correlations of water deficits
and radial growth in young red pine. Ecology 48: 525-530.

Carlson & Knight: Agrilus beetles 103

INDEX
Agrilus, genus, 3-4, 73, 90 nevadensis, 5-7, 80, 93
authorship, 1 niveiventris, see quadriguttatus
subgenera, 4-5, 73 olivaceoniger, 75, 77-78
synonyms otiosus, 4
Epinagrilus, 5 pensus*
Teres ,) populi, see granulatus
Uragrilus, 5 ~ politus, 4
Agrilus anxius group quadriguttatus, 12, 15, 81, 83184, 92-93
chromosome numbers of species, 2-3, 77 fulminans, 5-6, 75, 78, 86-87
definition, 4 hniveiventris, 5, 7, 78-80, 83-84, 86-87,
hybridization of species, 66, 77 92-93
ovipositor length of species 10, 76-77 quadriguttatus, 6, 75, 78, 84, 86-87,
phyletic tree, 75 92-93
Agrilus species, 21 rubicola communis, see aurichalceus
acutipennis, 74-75, 77-78 ruficollis, 6-7, 24, 73-77, 93
anxius* satoi, 75, 82
arcuatus, 4 | Sayi, 6
ater,” 73, °75,81, 92 solieri, 69
juriceki, 75, 81 semiaurovittatus, 75, 82
audax, 74-75 subcuneiformis, 75, 82
aurichalceus, 21 suensoni, 73
- auriventris auriventris, 40 suvorovi, 27
benjamini, 75, 78 tamanukii, 75, 81-82, 84
betulae, 1, 3, 5, 6, 15, 93 torpidus, 7-8, 37 :
bilineatus, 2, 4, 74-75, 78 trichocarpae, 6, 87
blanchardi, 5, 7 tscherepanovi, 75, 81
browni, 5, 7, 73-77, 84, 92, 93, 95 viridis, 1, 3-4, 17, 28-29, 41
champlaini, 73-75 populnea, see suvorovi
cinctus, 69 subspecies, in general, 4
criddlei, 8, 75, 78-79, 84, 92, 95 vittaticollis, 20-22, 36, 69, 74-77
fleischeri, 73, 75 yamawakii, 75, 78, 80
coreicus, 75, 81 Hosts of Agrilus species
kurosawai, 75, 81 alder, black, 16
nipponicola, 75, 81 alders, 81, 88
fulminans, see quadriguttatus Alnus spp., 3, 31, 78
granulatus, 19, 31, 85 crispa, 15, 31
granulatus, 6, 12, 15, 19, 81, 75, 82, rubra, 87-88
85-86 rugosa, 15-16, 19, 31, 37
liragus* angiosperms, 3
populi;, 6,75, 80-88 92 aspens, hybrid, 3, 15
eravis, 5, 8 aspen suckers, 2-3, 15, 17, 30-21, 32-36,
suerini, 73, 75, 79-81, 92 43-55, 92
horni* bayberry, 7; see also Myrica pensylvanica
lateralis, 5-7, 93 Betula spp., 2, 80
liragus, see granulatus alba, 15

*See table of contents, where each of the four principal taxa are sometimes listed by
their common names: the alder-birch borer (pensus), the aspen root girdler (horni),
the bronze birch borer (anxius), and the bronze poplar borer (granulatus liragus).

104

Contrib. Amer.

alleghaniensis, see lutea
fontinalis, 15

lenta, 15
lutea, 2, 14, 41
nigra, 15, 31

occidentalis, 15
papyrifera, 1-2, 14, 15, 18, 41, 63
var. commutata, 15
populifolia, 15
Betulaceae, 81, 87
birch
cut-leaf, 1
European white, 1, 14
river, 16-17, 88
white, 2
birches, 88
paper, 89
Carpinus caroliniana, 74

Comptonia peregrina, see Myrica asplen-

ifolia
Fagus sylvatica, 28
Myrica spp., 74
asplenifolia, 7
carolinensis, see pensylvanica

pensylvanica, 6, 74
oak, 78
Ostrya virginiana, 15-16, 74
poplar, 2, 15, 65
definition, 1
lombardy, 88
sprouts, 8, 77
Populus spp., 1-2, 44, 80, 85
alba, 2, 15
balsamifera
ssp. balsamifera, 15
ssp. trichocarpa, 15
deltoides, 15, 85
erandidentata, 3, 15, 32, 41
hybridization among, 85
laurifolia, 81
tremula, 15
tremuloides, 2, 15, 32-33, 35-36, 41,
43-46, 49-53, 55, 85
trichocarpa, see balsamifera
Quercus spp., 78
Roseaceae, 39
Rubus spp., 21, 74
Salicaceae, 81
Salix spp., 21, 74
lasiandra, 78
lasiolepis, 78
nigra, 78

Ent. Inst., vol. 4, no. 3, 1969

Ulmus rubra, 76

willows, 28
Invertebrates other than Agrilus spp.
Acmaeodera, genus, 90
ants, 37
Arthrochlamys, genus, 90
Atanycolus charus, 38
Buprestis sp., 56
Calligrapha, genus, 90
Carabidae, 83, 89
Carabus, European, 89
Cerambycids, 84
Chrysobothris sp. , 62

debilis, 56
Coccidencyrtus, sp. near, 83
Coccinellidae, 83
Dicerca sp., 62
Dolichomitus sp., 38

messor perlongus, 38
Doryctes atripes, 38
Ephialtes sp., 38
Eupristocerus cogitans, 16
Eurytoma magdalidus, 40
Gambrus extrematus, 40
Glypta sp., 37
Helconidea ligator, 38
Ichneumon sp., 38
Lampoterma sp., 40
Malacossoma disstria, 27
Mastrus smithii, 39
Oberea shaumii, 38
Olesicampe sp. , 38
Oobius sp., 39
Phasgonophora sulcata, 38
Pimplopterus sp., 37
Saperda spp.

calcarata, 38, 43

concolor, 38

moesta, 39
Spathius simillimus, 38
spiders, 37
Tetrastichus spp.

nordi, 39

rugglesi, sp. probably near, 38
Thysanus sp., 39
Xylophrurus spp.

agrili, 39

bicolor

bicolor, 39
maurus, 39

Structures of Agrilus spp.

abdominal groove of males, 4, 78, 81-82

Carlson & Knight: Agrilus beetles 105

aedeagus, 10-12, 55-56, 58-61, 64-65,

67, 80-81
anal forceps of larva, 13, 17
antennae, 66-69
antennal sockets, 67, 69
basal piece, 55
bursa copulatrix, 58
chorion, 13
common oviduct, 58
coxite baculum, 57
cuticle, 83 |
elytra, 11-12, 14, 76-81
elytral suture, 22
eyes, 22
genal ridge, 70
gonopore, 61
internal sac, 58-59, 62, 67
labial palpus of larva, 13, 14
labium of larva, 13, 14
lateral lobes, 55
lateral muscular ridges of larva, 17
mandibles, 22
marginal sclerites, 4, 11, 59-61, 74,
76-77
median lobe, 55
Ovipositor, 10, 56-57, 64, 76-77
Ovipositor tube, 56, 58, 61

parameres, 10-11, 55-56, 60

pedalis structuralis of larva, 13, 14

penis, 55-56, 58, 67, 78-79

prehumeral carinae, 74, 76-68

proctiger, 57

proctiger baculum, 57

prothorax of larva, 13

prosternal lobe, 70-71

pygidium, 9, 60

pygidial carina, 4

pygidial spine, 4, 9-10, 56, 58-61, 73, 76,
80 |

receptacle for penis apex, 58

rectum, 57

scape, 67, 69

scutellum, 66

spermatheca, 58

spiculate area of vagina, 58

spiracles of larva, 13

spiracular valve of larva, 13

stylets, 57

tarsal claws, 8-10, 73-74, 77

tegmen, 55-56, 65

unguifer, 9

vagina, 58

valvifer baculum, 57

vulva, 57-58

yaya ey rea
eet ae ;

AS
is

SAY
panes

-

Pai
me Pe

ae

4

Fase

i
ne
Tes
‘ le
a

(Continued from inside front cover)
Parts of Volume 4, with prices

No. 1. Mosquito studies (Diptera, Culicidae), 1968.
X. Peters, T. Michael. Dixinae originally described from North America. 7 pages.
XI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann. Mosquitoes
originally described from Argentina, Bolivia, Chile, Paraguay, Peru, and Uruguay.
29 pages. Price: $1.25, postpaid.

No. 2. Mosquito studies (Diptera, Culicidae), 1969.
XII. Berlin, O. G. W. A revision of the Neotropical subgenus Howardina of Aedes.
190 pages, 58 figures. Price: $6.00, postpaid.

No. 3. Carlson, Robert W. and Fred B. Knight. Biology, taxonomy, and evolution of four
sympatric species of Agrilus beetles (Coleoptera: Buprestidae). 105 pages, 47 figures.
Price: $3.50, postpaid.

Contributions
of the

American Entomological Institute

Volume 4, Number 4, 1969

MOSQUITO STUDIES (Diptera, Culicidae)
XIII. Pupae of the genus Psorophora in America North of
Mexico. By Ralph A. Barr and Sylvia Barr.

XIV. Redescription of Toxorhynchites (T.) gigantulus from
the Philippines. By O. G. W. Berlin.

XV. A new species of treehole breeding Anopheles from
the southwestern United States. By Thomas J. Zavortink.

CONTRIBUTIONS
of the
AMERICAN ENTOMOLOGICAL INSTITUTE

The Contributions are for larger papers on insects. Each paper is a separate number, with
separate pagination and index. Separate numbers aggregating about 500 pages constitute a volume.
Issues appear irregularly, as suitable manuscripts are available.

Copies are sold separately or in subscriptions to complete volumes. Complete volumes are
$12.00. The price of separate numbers varies. Subscribers are billed for each volume with its
beginning number, and receive the parts as issued. Orders for separate numbers that total less
than $8.00 must be accompanied by payment. Address orders or correspondence to the American
Entomological Institute, 5950 Warren Road, Ann Arbor, Michigan 48105, U.S.A.

Parts of Volume 1, with prices

No. 1. Dasch, Clement E., 1964. The Neotropic Diplazontinae (Hymenoptera, Ichneumonidae).
75 pages, 69 figures. Price: $2.25, postpaid.
No. 2. Mosquito studies (Diptera, Culicidae), 1965.
I. Belkin, John N., etal. A project for a systematic study of the mosquitoes of
Middle America.
Il. Belkin, John N., et al. Methods for the collection, rearing and preservation of
mosquitoes.
78 pages, 4 figures. Price: $2.25, postpaid.
No. 2a. Same as no. 2, but in Spanish. Price: $2.25, postpaid.
No. 3. Matthews, Robert W., 1965. The biology of Heriades carinata Cresson. (Hymenoptera,
Megachilidae). 33 pages, 23 figures. Price: $1.00, postpaid.
No. 4. Mosquito studies (Diptera, Culicidae), 1965.
Ill. Ramalingam, Shivaji, and John N. Belkin. Two new Aedes from Tonga and Samoa.
IV. Belkin, John N. The mosquitoes of the Robinson-Peabody Museum of Salem
expedition to the southwest Pacific, 1956.
34 pages, 3 figures. Price: $1.00, postpaid.
No. 5. Mosquito studies (Diptera, Culicidae), 1965.
V. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.
Mosquitoes originally described from Middle America. 95 pages. Price: $2.75 postpaid.
No. 6. Mosquito studies (Diptera, Culicidae), 1966.
VI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.
Mosquitoes originally described from North America. 39 pages. Price: $1.00, postpaid.
No. 7. DeLong, Dwight M. and Paul H. Freytag, 1967. Studies of the world Gyponinae
(Homopotera, Cicadellidae). A synopsis of the genus Ponana.
86 pages, 257 figures. Price: $2.50, postpaid.
No. 8. Contributions to the mosquito fauna of southeast Asia, 1967.
I. Delfinado, Mercedes D. The genus Aedes, subgenus Neomacleaya Theobald in
Thailand. 55 pages, 20 figures. Price: $1.75, postpaid.

Parts of Volume 2, with prices

‘ No. 1. Contributions to the mosquito fauna of southeast Asia, 1967.
Il. Bram, Ralph A. The genus Culex in Thailand (Diptera: Culicidae).
296 pages, 104 figures. Price: $9.00, postpaid.
No. 2. Schuh, Toby, 1967. The shore bugs (Hemiptera: Saldidae) of the Great Lakes region.
35 pages, 30 figures. Price: $1.25, postpaid.
No. 3. Snider, Richard J., 1967. The chaetotaxy of North American Lepidocyrtus s. str.
(Collembola: Entomobryidae). 28 pages, 57 figures. Price: $1.00, postpaid. —
No. 4. Contributions to the mosquito fauna of southeast Asia, 1968.
Ill. Delfinado, Mercedes D. The genus Aedes, subgenus Neomacleaya Theobald
of southeast Asia. 76 pages, 26 figures. Price: $2.25, postpaid.
No. 5. Contributions to the mosquito fauna of southeast Asia, 1968.
IV. Knight, Kenneth L. Species of the subgroup Chrysolineatus of group D, genus
Aedes, subgenus Finlaya Theobald. 45 pages, 12 figures. Price: $1.50, postpaid.

Parts of Volume 3, with prices

No. 1. Mosquito studies (Diptera, Culicidae), 1968.
VIl. Belkin, John N. The Culicidae of New Zealand.
182 pages, 30 figures. Price: $5.50, postpaid.
No. 2. Mosquito studies (Diptera, Culicidae), 1968.
VIll. Zavortink, Thomas J. A prodrome of the genus Orthopodomyia.
221 pages, 41 figures. Price: $6.75, postpaid.
No. 3. Fleming, Richard C., 1968. Head musculature of sphinx moths (Lepidoptera:
Sphingidae). 32 pages, 37figures. Price: $1.00, postpaid.
No. 4. Mosquito studies (Diptera, Culicidae), 1968.
IX. Belkin, John N. The type specimens of New World mosquitoes in European
museums. 69 pages. Price: $2.00, postpaid.

(Continued on back cover)

MOSQUITO STUDIES (Diptera, Culicidae)

XIII. PUPAE OF THE GENUS PSOROPHORA IN

NORTH AMERICA AND PUERTO RICO!
By

A. Ralph Barr?
and |

Sylvia Barr

Fourteen species of Psorophora have been reported from America north of Mexi-
co and Puerto Rico (Carpenter and LaCasse 1955; Stone, Knight and Starcke 1959).
It is not possible at the present time to identify pupae in this genus because none
has been described in detail. Neither is there a published key by which pupae of this
genus may be separated from others occurring in North America. This study con-
tains descriptions of 13 of the 14 species of Psorophora known from North America
and Puerto Rico and discusses their identification. A generic diagnosis and discus-
sion of generic classification is also included.

This study was made possible by the loan or donation of specimens by the fol-
lowing individuals: Harry D. Pratt, Alan Stone, John N. Belkin, James S. Haeger and
Nina Branch, Osmun P. Breland, R.E. Bellamy, W.R. Horsfall, William Wills, and
Harold C. Chapman. To all of these individuals the authors are deeply grateful.

MATERIALS AND METHODS

Pupal exuviae were mounted as described by Barr (1958). After the specimen was
dissected and positioned in a small amount of thin Canada balsam, a large coverslip
was placed on the exuviae and gentle but firm pressure applied to flatten the prep-
aration. Specimens were examined with a phase contrast microscope at an appropri-
ate magnification, usually 128 diameters, for counting branches of setae. The num-
bering system used for the abdominal setae was that of Barr and Myers (1962).
Cephalothoracic setae were numbered in the conventional manner as illustrated in
Barr (1963). The ‘‘usual’? number of branches as used in this study is a restricted
range which usually includes 90% or more of the observations; only unusual values

1Contribution from the Division of Infectious and Tropical Diseases, School of Public Health,
UCLA, and from the project “Mosquitoes of Middle America,” supported by U.S. Public Health
Service Research Grant AI-04379 and U.S. Army Medical Research and Development Command
Research Contract DA-49-193-MD-2478.

2School of Public Health, University of California, Los Angeles, California 90024.

2 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

are excluded. The widths of respiratory trumpets as given must be interpreted with
care as they vary with mounting method and are not strictly comparable in various
kinds of mounts.

Genus Psorophora Robineau-Desvoidy, 1827

PUPA. Very large to small. Cephalothoracic setae more or less normal for the family; 1 long;
2 usually shorter than 3; 4 and 5 well developed and subequal; 7 much longer than 6; 8 and 9
usually well developed and well separated. Respiratory trumpet about 1-1/2 to 4 times as long as
its pinna. Abdomen usually without reticulation between float hairs (1) on I; float hairs usually
with small number of stalks at base; 2 usually medial of 4 on I; I-6 and I-7 subequal in length,
both usually branched; I-9 commonly branched; venter of I commonly with a seta. Seta 5 usually
longer than, and medial of, 3 on II; 6 and 7 usually subequal in length and branched on IJ; venter
of If commonly with a “transitory” seta. Seta 5 usually longer than, and medial of, 3 on III.
Setae 1 and 5 usually well developed and subequal in length on IV and V; 4 usually longer than,
and medial of, 3 on these segments. Seta 3 medial of, and occasionally longer than, 1 on VI; both
usually much shorter than VI-5. Seta 6 usually smaller than, and anterior and medial of, 9 on VII.
Segment VIII usually with conspicuous extensions (lobes) of the sternum and tergum which par-
tially cover the bases of the paddles (fig.1a); seta 9 inserted on the venter slightly before the pos-
terior corner, occasionally on the margin. IX without a seta. Paddles pear-shaped, with conspicu-
ous midrib not reaching margin; usually without strong denticulation; accessory seta present or
absent; paddle seta commonly branched.

There appears to be no single character by which the genus may be defined. The
conspicuous flaps on VIII (fig.la) have not been seen by the writers in pupae of
other genera but, unfortunately, are not well developed in the subgenus Grabhamia
(fig.2a). The position of 6 on VII (anterior of 9) is unusual although not invariable
in the genus. The placement of 3, medial of 1, on VI is unusual but not unique to
the genus. Another difficulty with this character is that 3 and 1 are not always read-
ily distinguishable. If 1 is much larger than, but lateral of 3, as is usually the case,
the reversal is quickly noted. If, on the other hand, 3 is as large as or larger than 1,
the reversal is not apparent. These setae seem always to be reversed in Psorophora
pupae, however, as shown by an examination of the prepupa or larva (Barr and
Myers 1962) in doubtful cases. The reduction in the seta which is apparently VI-1
(actually 1 or 3) as compared with 5 is characteristic of the genus. The reduction in
size of the dorsal setae on VII is also characteristic of, although not unique to, the
genus. 7

Psorophora pupae are most likely to be confused with those of Culiseta or Aedes.
Pupae of Psorophora have either: (1) large spines in the posterior corner of segment
IV (fig.6), (2) an accessory paddle seta, or (3) prominent lobes on the posterior mar-
gins of segment VIII. These characteristics do not occur in Aedes or Culiseta pupae
except that an occasional species, as Culiseta melanura (Coquillett) or Aedes atro-
palpus (Coquillett), may have an accessory paddle seta. The following key will aid
in separating these groups.

KEY FOR SEPARATION OF PUPAE OF
PSOROPHORA, CULISETA, AND AEDES

1. Posterior corner of segment IV with large spines (fig.9).

Barr and Barr: Psorophora Pupae 3

. Psorophora (Janthinosoma) part

Posterior'c corner of segment IV without ine eninge .6. 82 20 Rie
Posterior margin of sternum VIII with conspicuouslobes ....... .3
Posterior margin of sternum VIII without conspicuouslobes . ..... .4
Accessory paddle seta presents). 2. Psorophora (Psorophora)
Accessory paddle seta absent . . . . . . .Psorophora(Janthinosoma) part
Accessory paddle setaprovent a) 2 6.90). ke CE Oe we ee ed
Accesséty paddle seta absent! :0) 2) ye8e ae 2 OF ee a GO ee ee

Either respiratory trumpet split almost completely to base, or paddle margin
with large, needle-like denticles, or paddle deeply incised; 9 anterior of 6 on
WER occ ames . 6

Respiratory trumpet not split almost completely to base: paddle margin with-
out large, needle-like denticles; paddle not deeply incised: 9 usually not an-
benign ot GOON Hes aw PO ee _Psorophora (Grabhamia)

Setae 1 and 5 subequal in length on IV to VI or paddle margin with long,
needle-like denticles; with seta on IX; with prominent spines on hypopygium

. . Culiseta

Seta 5 usually much longer than A on TV to VI: paddle margin usually without
long, needle-like denticles; usually without seta on IX; usually without prom-
inent spines On hypopyeiwm :\°. 6 a ee ee ee ee Aedes

KEY TO NORTH AMERICAN PSOROPHORA PUPAE

Posterior corners of IV with large spines (fig.9); accessory paddle seta usually
absent VJanthinosoma, except cyanescens). . . een

Posterior corners of IV without large spines ventrally; accessory paddle seta
usually present (except cyanescens), sometimes indistinct .

CT-10 usually with 3 or fewer branches; II-5 usually with 4 or fewer branches; |

III-1 usually with 6 or fewer branches; III-5 usually with 4 or fewer branches

.varipes

CT- 10 usually with 5 or more branches: IL 5 usually with 5 or more branches;
III-1 usually with 7 or more branches; III-5 usually with 6 or more branches

a

Heavily pigmented, grossly appears brown, not yellow; respiratory trumpets less
than 3 times as long as wide; CT-12 usually with 5 or more branches; VII-1
frequently with 3 or more branches . ‘ai 2a pegohnstonii

Lightly pigmented, grossly appears yellow; respiratory trumpet 3 or more times
as long as wide; CT-12 usually with 4 or fewer branches (except horrida);
Vibi sinter gouge 6. ye ee a es eee ie ee

II-1 usually with 5 or fewer branches; I-2 usually single; V-8 usually double or
single; VI-11 usually single; male terminalia asin fig.6. . . . . longipalpus

10.

Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

II-1 usually with 7 or more branches; I-2 usually with 2 or more branches; V-8

usually with 3 or more branches; VI-11 usually branched; male terminalia as
Te as a ee ee a oe ye oS

CT-12 usually with 2 or fewer branches; I-4 usually with 4 or fewer branches;

II-4 usually with 3 or fewer branches . . . . sacs ferox
CT-12 usually with 4 or more branches; I-4 usually with 5 or more branches;
II-4 usually with d-ormore branches... .- g..0b ae, Bok oo eesborrida

Pupa very large; segment III as in fig.1, 4 usually closer to 5 than to 1, usually
anterior of and subequal to 5 in length; II-5 and III-5 usually with 3 or fewer
branches; VIII-9 usually with 5 or fewer principal branches; CT-10 usually
with 4 or fewer branches; CT-8 usually single (Psorophora). . . el:

Pupa usually medium to small: segment III as in fig.2, 4 usually closer to ] than
to 5, usually posterior of and much longer than 5; Tl- 5 and III-5 usually with
4 or more branches; VIII-9 usually with 6 or more principal branches; CT-10
usually with 5 or more branches; CT-8 usually with 2 or more branches. . 8

Segment VIII with dark spots anteriorly and posteriorly on either side of me-
dial line, entire segment usually spotted; without dark medial stripe on abdo-
men; with diagonal stripes enclosing conspicuous light areas laterally on V
to VII (fig.4) . . . «= ROWardE

Segment VIII without such dark spots anteriorly and ‘posteriorly although usu-
ally with darker medial stripe, segment usually rather evenly dark laterally
and on mid-line, with light pigmentation elsewhere; with dark medial stripe
on abdomen; without pattern as described above although with diagonal
stripes laterally on segments V to VII Giz.3) . 6°. SS SD iiata

Accessory paddle seta usually absent; CT-11 usually with 3 or more branches;
I-4 usually with 6 or more branches; II-4 usually with 4 or more branches;
III-4 usually with 4 or more branches . . . . eyanescens

Accessory paddle seta usually present; CT-1 1 ‘usually single, occasionally dou-
ble; I-4 usually with 4 or fewer branches; II-4 usually with 3 or fewer
branches: III-4 usually single or double (Grabhamia) oe ee eee

VII-6 with 4 or fewer branches, usually double or triple; respiratory trumpet
usually 2 or fewer times as long as its pinna; I-3 usually with 10 or fewer
branches; V-3 with 5 or fewer branches .. . . . . insularia

VII-6 with 6 or more branches; respiratory trumpet usually ¢ over twice as long
as its pinna; I-3 usually with more than 10 branches; V-3 usually with 6 or
fidke Bi bies “Oceans oo Sot on eke a eS age

I-5 with 4 or fewer branches; I-7 with 5 or more branches; III-6 with 3 or more
branches; [V-1 with 4 or more branches; IV-6 with 3 or more branches; V-1
usually branched; V-8 with 4 or more branches; VI-6 frequently with 3 or
more branches; VII- 5 with 3 ormore branches . . . . pygmaea

I-5 with 5 or more branches; I-7 usually with 4 or fewer branches: IlI- 6 usually
double or single; IV-1 double or single; IV-6 double or single: V-1 usually
single (except discolor); V-8 usually with 3 or fewer branches; VI-6 single or
double; Vil-S: usuaiy double-or single fat eh eee dey wee 1

Barr and Barr: Psorophora Pupae 5

11. I-4 with 4 or more branches; VII-3 usually with 4 or more branches
Sl a ic tle i a Ae wind a Ar alae dl baa Us ele idl ca ati ag a . signipennis
I-4 single or double; VII-3 usually with 3 or fewer branches. . . ... . 12
12. Paddle seta very long, about 1/4 or more the length of the paddle (fig.8); VII-1
usually placed about midway between lateral edge of segment and mid-line
AN LTRS SES SRY US SENSES TP TIMER ERO EE SAAR Hoe ROR CH ge ee pee got RR aoe. Oe EN
Paddle seta of normal length, usually less than 1/6 the length of the paddle;
VII-1 usually much closer to mid-line than to lateral edge of segment (fig.2)
. confinnis

Subgenus Psorophora Robineau-Desvoidy, 1827

PUPA. Very large; with conspicuous dark areas, either in lines (ciliata) or spots (howardii), on
various parts of the abdomen; bases of major setae usually surrounded by dark spots; setation
generally short in comparison with that of the other subgenera. CT-8 usually single; CT-10 usually
with 4 or fewer branches; respiratory trumpet usually less than twice as long as its pinna. II-5 usu-
ally with 3 or fewer branches. Seta 4 removed laterally, and similar to 5, on II and III; usually it
is at least as close to 5 as to 1 and only slightly, if at all, posterior of 5. Usually with a “transitory”
seta or at least a conspicuous sensillum on the venter of II. III-1 usually with 4 or fewer branches;
III-5 usually with 3 or fewer branches. VI-3 usually smaller than VI-1 and both conspicuously
smaller than VI-5. Segment VIII with conspicuous lobes on posterior margin of sternum; VIII-9
usually with 5 or fewer branches. Accessory paddle seta present although frequently indistinct;
paddle usually obliquely truncate and about as wide as long.

Pupae of the subgenus Psorophora can usually be easily recognized by their large
size, extensive pigmentation, wide paddles, presence of accessory paddle seta, pres-
ence of lobes on sternum VIII, and pattern of setation on II and III.

Psorophora (Psorophora) ciliata (Fabricius, 1794)

PUPA. Not previously described in its entirety. Partial descriptions have been given by Mitchell
(1907), Howard, Dyar, and Knab (1912, 1917), Knight and Chamberlain (1948), and Lane (1953).

Very large; respiratory trumpet 1.8 to 2.1 times as long as its pinna; occasionally with a ventral
seta on I. Venter of II usually with conspicuous sensilla. Paddle about 1.2 to 1.3 times as long as
wide, with few, small, rounded denticles. A summary of the setal branchings is given in Table 1.

This pupa is not likely to be confused with others except howardii. The abdom-
inal markings have been described by Lane (1953) and, although variable, seem in-
variably to be reliable for separating this pupa from that of howardii. The exuviae
of ciliata grossly appear to have three longitudinal, dark lines while those of howardii
have a mottled appearance. CT-6 may also be useful in this separation. P. howardii
also generally has broader paddles with more and longer denticles than does ciliata.

Material examined: 3 specimens from Virginia, 2 from Maryland, and 1 from New
York, all in the U.S. National Museum; 19 specimens from St. Paul, Minnesota in
the collection of the writers.

6 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

Psorophora (Psorophora) howardii Coquillett, 1901

PUPA. Not previously described in its entirety although partial descriptions are given by Mitch-
ell (1907) and Lane (1953). The abdominal markings have been described by Lane (1953).

Large to very large; abdomen conspicuously spotted; respiratory trumpet 1.7 to 2.0 times as
long as its pinna; usually with a ventral seta on I; usually with a “transitory” seta on the venter of
II; paddle 1.0 to 1.1 times as long as wide, with moderate number of small to medium, rounded
denticles. A summary of the setal branchings is given in Table 1.

Material examined: 5 specimens from Louisiana, 1 each from Arkansas and Mis-
souri, 2 from Mexico, | from Nicaragua, and 1 without data, all in the U.S. National
Museum. Also 16 specimens from Florida in the collection of the writers, supplied

by J.S. Haeger and Nina Branch.

Subgenus Janthinosoma Lynch Arribalzaga, 1891

PUPA. Usually medium or small; CT-8 usually branched; CT-10 usually with 5 or more branches
(except varipes); respiratory trumpet about 2 to 4 times as long as its pinna. II-5 usually with 4
or more branches. Seta 4 closer to 1 than to 5 on III and frequently on II, usually markedly longer
and much posterior of 5. III-1 usually with 5 or more (except cyanescens) and III-5 usually with
4 or more (except varipes) branches. All species except cyanescens have a patch of large, cuticular
spines in the posterior corner of IV. On VI seta 3 may be larger or smaller than 1. Segment VIII
with conspicuous lobes on posterior margin of sternum; VIII-9 usually with 6 or more branches.
Accessory paddle seta absent. Paddle with weak denticulation.

Janthinosoma pupae (except cyanescens) may be easily recognized by the large
spines on IV; these do not occur in other subgenera. The subgenus is also distin-
guished by lacking an accessory paddle seta although it occasionally occurs as an
anomaly in at least johnstonii and varipes.

Psorophora (Janthinosoma) cyanescens (Coquillett, 1902)

PUPA. Previously undescribed. Medium to large; respiratory trumpet 1.7 to 2.6 times as long
as its pinna; frequently with a ventral seta on I; occasionally with a “transitory” seta on II. A
summary of the setal branchings is given in Table 1.

The pupa of this species has II-4 more highly branched than in other Psorophora
species. It can be separated from Grabhamia and Psorophora by lacking the acces-
sory paddle seta and from other Janthinosomas by lacking the prominent spines on
IV.

Material examined: 10 specimens from Wilson Dam, Alabama, in the collection of
J.N. Belkin (UCLA); and 11 specimens from Cameron, Louisiana, in the collection
of the writers, supplied by H.C. Chapman.

Psorophora (Janthinosoma) ferox (Humboldt, 1819)

PUPA. Partially described by Mitchell (1907), Howard, Dyar, and Knab (as sayi; 1912, 1917)
and Lane (1953).

Barr and Barr: Psorophora Pupae 7

Medium to large; respiratory trumpet 3.0 to 4.3 times as long as its pinna; frequently with a
ventral seta on I; paddle with sparse, weak denticles. A summary of the setal branchings is given
in Table 1.

The presence of large spines on IV marks this species as a typical Janthinosoma,
and allies it with horrida, longipalpus, johnstonii, and varipes. It is generally marked-
ly larger than the other four species; it also has longer respiratory trumpets and less
pigmentation than johnstonii. The key characters given should allow the separation
from varipes without difficulty. The pupa of ferox is most likely to be confused with
those of longipalpus and especially horrida; the key as given serves fairly well for the
specimens at hand but may require modification when additional material is exam-
ined.

Material examined: 2 specimens from Piraja, Brazil in the collection of the US.
National Museum; 1 specimen without data in the collection of R.E. Bellamy; 9
specimens from Wilson Dam, Alabama, in the collection of J.N. Belkin (UCLA); and
4 specimens from Lufkin, Texas, collected by O.P. Breland and 26 specimens from
Florida, collected by J.S. Haeger, in the collection of the writers.

Psorophora (Janthinosoma) horrida (Dyar and Knab, 1908)

PUPA. Previously undescribed; small to medium; respiratory trumpet 3.0 to 3.7 times as long
as its pinna; no ventral seta on I or II in the specimens examined. Paddles with sparse, weak den-
ticulation. A summary of the setal branchings is given in Table 1.

This pupa has CT-12, I-9, II-4, III-4, I-10, VI-4 and VI-8 uncommonly highly
branched. The description is based on 15 entire pupae from Illinois reared by W.R.
Horsfall. Additional characteristics may become evident when exuviae are examined.

Psorophora (Janthinosoma) longipalpus Randolph and O’Neill, 1944

PUPA. Previously undescribed; small to medium; respiratory trumpet 3.0 to 3.8 times as long
as its pinna; frequently with a ventral seta on I; venter of II usually without a “transitory” seta;
paddle with few, small denticles. A summary of the setal branchings is given in Table 1.

CT-10 and II-1 have fewer branches than in most species. The pupa is most likely
to be confused with those of horrida and ferox. The bulbous terminalia of the male
are evidenced in the pupa by the very wide basimeres.

Material examined: 12 specimens from Douglas County, Kansas, in the collection
of the writers.

Psorophora (Janthinosoma) johnstonii (Grabham, 1905)

PUPA. Partially described by Thurman, Haeger, and Mulrennan (1951) and Lane (1953).

Small; with dark, rather uniform pigmentation; respiratory trumpet 2.3 to 2.8 times as long as
its pinna; ventral seta frequently present on I; II frequently with “transitory” seta; paddle with
very few, small denticles; accessory paddle seta occasionally present. A summary of the setal
branchings is given in Table 1.

8 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

The pupa of this species is a small, dark Janthinosoma with short respiratory
trumpets. It has CT-12 uncommonly highly branched for this group.

Material examined: 12 specimens from Florida collected by J.S. Haeger in the
collection of the writers.

Psorophora (Janthinosoma) varipes (Coquillett, 1904)

PUPA. Partially described by Mitchell (1907).

Small; respiratory trumpet 2.1 to 3.9 times as long as its pinna; usually with ventral seta on I
and “transitory” seta on II; paddle with medium to large denticles; accessory paddle seta occa-
sionally present. A summary of the setal branchings is shown in Table 1.

This is a small Janthinosoma pupa with generally reduced setal branchings. It can
be distinguished from related forms especially in the reduced branching of CT-10,
I-3, I-4, 1-5, H-5, TI-5, V-7, and VI-6.

Material examined: 16 specimens from Wilson Dam, Alabama, in the collection of
J.N. Belkin (UCLA); 1 specimen from Georgia in the collection of R.E. Bellamy.

Subgenus Grabhamia Theobald, 1903

PUPA. Usually medium to small; CT-8 usually branched; CT-10 usually with 5 or more
branches; respiratory trumpet about 1-1/2 to 3 times as long as its pinna. II-5 usually with 4 or
more branches. Seta 4 usually closer to 1 than to 5 on III and frequently on II, usually markedly
longer and much posterior of 5. III-1 usually with 5 or more and III-5 with 4 or more branches.
Seta 1 is usually larger than 3 on VI. Segment VIII without conspicuous lobes on posterior mar-
gin of sternum; VIII-9 with 6 or more branches. Accessory paddle seta present. Paddle usually
with weak denticulation.

Grabhamia pupae lack the conspicuous lobes on VIII of the other two subgenera,
but possess an accessory paddle seta. They are most likely to be confused with Culi-
seta and Aedes pupae which, however, generally lack an accessory paddle seta. In
addition, Grabhamias lack the seta on IX found in Culiseta and occasionally in Ae-
des. The position of 6 with respect to 9 on VII is also useful in distinguishing Grab-
hamia pupae from those of Aedes and Culiseta. :

Psorophora (Grabhamia) confinnis (Lynch Arribalzaga, 1891)

PUPA. Partially described by Mitchell (1907), Howard, Dyar, and Knab (as columbiae; 1912,
1917), and Lane (1953).

Medium to large; respiratory trumpet 1.9 to 2.8 times as long as its pinna; frequently with ven-
tral seta on I; II occasionally with “transitory”’ seta; paddle with few, small denticles. A summary
of the setal branchings is given in Table 1.

The pupa of this species is characterized by lacking characteristics of other Grab-
hamia pupae. Two specimens from the Virgin Islands were excluded because of the
highly branched nature of CT-4, 10, 11, seta 6 on I to III and VI, IV-1, and VII-5.
It is problematic whether these specimens are conspecific with the others. Three
specimens from Brazil and two from Mexico agreed reasonably well with material

Barr and Barr: Psorophora Pupae 9

from the United States.

Material examined: 3 specimens from Porto Alegre, Brazil; 2 specimens from Sa-
lina Cruz, Mexico; 3 specimens from Arkansas, 2 from Louisiana, 3 from Virginia,
2 from St. Croix, Virgin Islands, and 4 without data, all in the U.S. National Muse-
um. Also 16 specimens from Florida, collected by J.S. Haeger, and 25 from Thermal,
California, collected by William Wills in the collection of the writers.

Psorophora (Grabhamia) discolor (Coquillett, 1930)

PUPA. Partially described by Mitchell (1907).

Medium size; respiratory trumpet about twice as long as its pinna; frequently with ventral seta
on I; occasionally with “transitory” seta on II; paddle with few, small denticles. A summary of the
setal branchings is given in Table 1. |

The pupa of this species may be distinguished from those of all others in the
genus by the extremely long paddle seta which is generally more than 1/4 the length
of the paddle; in related species it is generally less than 1/6 the length of the paddle.
In addition, setae 1 are frequently displaced laterally on V and VII. The exception-
ally long paddle seta was noted by Mitchell (1907).

Material examined: 6 specimens from Eagletown, Oklahoma, and 1 without data
in the collection of the U.S. National Museum; 13 specimens from Wilson Dam, Ala-
bama, and one from Pickwick Lake, Mississippi, in the collection of J.N. Belkin
(UCLA); and 3 specimens from Johnson County, Kansas, in the collection of the
writers.

Psorophora (Grabhamia) signipennis (Coquillett, 1904)

PUPA. Paddle and trumpet figured by Yamaguti and LaCasse (1951).

Medium size; respiratory trumpet 1.5 to 1.7 times as long as its pinna; without a ventral seta
on I or a “transitory” seta on II. Paddle with sparse, small denticles. A summary of the setal
branchings is given in Table 1.

The pupa of this species is similar to most Grabhamia pupae except for I-4 which
is exceptionally highly branched.

Material examined: 24 specimens from Poplar, Montana, in the U.S. National Mu-
seum, and 1 from Johnson County, Kansas, in the collection of the writers.

Psorophora (Grabhamia) insularia (Dyar and Knab, 1906)

PUPA. Previously undescribed; small; respiratory trumpet 1.3 to 1.9 times as long as its pinna;
2 usually lateral of 4 on I; often with a ventral seta on I and occasionally with a “transitory” seta
on II; denticles on paddle medium to large, sparse to abundant. A summary of the setal branchings
is given in Table 1.

This very distinct short-trumpeted pupa is characterized also by usually having 2
lateral of 4 on I, by the denticulation of the paddle, and by the small number of
branches in VII-6 and I-3.

10 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

The description is based on 3 exuviae, all male, from Losey Field, Puerto Rico,
December 1, 1942, in the collection of H.D. Pratt, and 3 pupae, all female, from
Mona Island, Puerto Rico, collected by J. Maldonado, April 16, 1954, in the collec-
tion of the U.S. National Museum. The material from Losey Field had been identi-
fied as pygmaea but differed conspicuously from other material identified as that
species from Catafio (see pygmaea). The identification as insularia was by compari-
son with Maldonado’s material supplied by Alan Stone. The two lots of insularia also
have many differences although both are clearly different from pygmaea; the Mona
Island material has more branching in CT-6 and 9, I-5 and 6, II-4 and 6, III-4 and 8,
IV-3, VII-8 and 11, and VIII-14 than does the Losey Field material. It also has longer
trumpets, much less prominent denticulation and only occasionally has 2 lateral of
4 on I. It remains to be seen whether or not these two types of insularia are conspe-
cific; for the present the similarities of the two justify their being considered con-
specific.

Psorophora (Grabhamia) pygmaea (Theobald, 1903)

PUPA. Previously undescribed; small; respiratory trumpet 2.6 to 3.1 times as long as its pinna;
usually with a ventral seta on I and occasionally with a “transitory” seta on II; paddle with small
denticles laterally. A summary of the setal branchings is given in Table 1.

This pupa can be distinguished from other Grabhamia pupae (except insularia)
by the small number of branches in I-5 and the large number in IV-1.

Material examined: 2 66 and 2 29 from Catafio, Puerto Rico, May 7, 1943, in the
collection of H.D. Pratt.

REFERENCES CITED

Barr, A.R.

1958. The mosquitoes of Minnesota (Diptera: Culicidae: Culicinae). Minn. Univ.
Agr. Exp. Sta., Tech. Bull. 228. 154 p.

1963. Pupae of the genus Culiseta Felt. II]. Descriptions and a key to the North
American species (Diptera: Culicidae). Entomol. Soc. Amer., Ann. 56:324-330.

Barr, A.R. and C.M. Myers

1962. Pupae of the genus Culiseta Felt. I. The homology of larval and pupal se-

tae (Diptera: Culicidae). Entomol. Soc. Amer., Ann. 55:94-98.
Carpenter, S.J. and W.J. LaCasse

1955. Mosquitoes of North America (north of Mexico). Berkeley and Los Ange-

les, Univ. Calif. Press. 360 p.
Howard, L.O., H.G. Dyar and F. Knab

1912(1913). The mosquitoes of North and Central America and the West Indies.
Volume II. Plates. Wash., Carnegie Inst. 150 pl. (Publication 159).

1917. The mosquitoes of North and Central America and the West Indies. Volume
IV. Systematic description. Part II. Wash., Carnegie Inst. p.525-1064 (Publi-
cation 159).

Knight, K.L. and R.W. Chamberlain

1948. A new nomenclature for the chaetotaxy of the mosquito pupa, based on a
comparative study of the genera (Diptera: Culicidae). Helminthol. Soc. Wash..,
Proc. 15:1-18.

Barr and Barr: Psorophora Pupae 11

Lane, J.
1953. Neotropical Culicidae. Vol. II. Sao Paulo, Univ. Sao Paulo. p.553-1112.
Mitchell, E.G.
1907. Mosquito Life. New York, Putnam. 281 p.
Stone, A., K.L. Knight and H. Starcke
1959. A synoptic catalog of the mosquitoes of the world (Diptera, Culicidae).
Wash., Entomol. Soc. Amer. (Thomas Say Found. Publication 6). 358 p.
Thurman, E.B., J.S. Haeger and J.A. Mulrennan
1951. The taxonomy and biology of Psorophora (Janthinosoma) johnstonii (Grab-
ham, 1905) (Diptera: Culicidae). Entomol. Soc. Amer., Ann. 44:144-157.
Yamaguti, S. and W.J. LaCasse
1951. Mosquito fauna of North America. Part II]. Genera Orthopodomyia, Man-
sonia, and Psorophora. Kyoto, Honshu, Japan Logistical Command, Office
Surg. 92 p. |

SV
cl
87S
2/0
aa é
ca
6-9

I
v1
Gt
oe

I
CA
VC
4
cl

vl
vl
LS
Ol-€
WA
6-€
Se
TC
ce
81-8
Ci
£7¢
Cd
ae
I
Sr
VI
can
us

mi Os hh BOS Ge ES 6-S L-p 8-¢ OI-9 LS
eh ge eo ea ee a L-€ v1 oh. t4 al
ee ep L-b OI-9 6-9 8-S ayo Ol..- 219
VW. Gee SE PL CL 81-8 91-6 lie S0r°. fs
CA ee ee 23) ee Ee 1 Fl cl ee €-]
2 Ho 84 oe fe $e L-v Sp oe 4 r-€
& Glewe iirc |, Sa Std G-€ in Sp 9 7
J ce a6 OP = 6S. TIS 01-9 g-¢ pe eee ae
a Ce Ore © ere os FZ Lis v7 Ch. oe ct
= TEL OCrE BCE. €o-91 9-1 CClL OPE. SIX Scr ea
if a oa fl 8ee Ce OE ue
s [ae fee ee eee, re owe ee ee
iE ot or = ee ec £1 v7 oe ch oc I
7 2 eo ae ees mee Ce. 86° 610 lcm oe
: | Vo Se a: 6 cs C-¢ oT €-7
2 ZI OET eee ee 0 ee CE bt OE
8 Ceo 9o ee aie Se 9-7 S-€ Lp Ob r-€
A he eee ee ae eg C-€ C-€ SS ee r-7
cl ae Z gd ages CBee =] é ZI
sadispa 11U0JS sndjod DpissOYy xo1af SUZISAUDAD DIADINSUl1 papusAd siuuad 40] OISI1p s1uu1fuor 1IPADMOY DID1}19
-uyol -18uo] -1U381S
(DUOSOUlYUDs ) (DIUIDYGDA4 )

12

(v1oydoios )

‘aednd evroydososg SUIAJI}USpI JOJ [NJosn avjOsS UI SoyoULIG Jo Joquinu [eNsy “| sIGe IL

Se ie ee I To © A a 2 WR I

N

N
on

=
=
i}

co 6O
ec ag

at SP a oO

II

LO

13

Barr and Barr: Psorophora Pupae

I

0 = = © + Vw VCO Ke CO ew = OH eS SO & CC Oy ee me er oS NO

= £43

= wr

Al

Ill

Il

Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

14

CS OFeE vo: tee 6S wy 119 Go? OFrL Clee OI-9 of St 6 “HA

ee | ‘a So £% vt es v1 ent v-C v-C cA Sa Me %:
ne. at ct eet C el I I I I I I fi. Bh
L.. Fc Ee To 6G ZI-S S-¢ L-v 6-7 a ov Oo he. a
ge ee fo. OF ft 6I-L S-¢ ee iis OI-¢ Ot-€ [as C9. OS
ct 6S oe Se Bees oe ae gs I I a4 I od I is
ee Vie. es SI-6 OTS LOT e"t era; Aa% LI-8 SI-L eS. FR
et C ot 14 ee cl Se v-C ge G (Gua | i ae
Se ye S-€ St. re es at 8 ro. fe v-C Ze ah te
CL ort cl ei te a a | CT a | Cl I ae os. 4774s

I I I Gi. et ag v-C S-C v-C ot S-C €-1 a eee &
ci 3 cc. oe Se L-V v-C Sv ge a S-€ eo Se SE

ty ott I cr. ot (ua | ae ct ot I C4 fie | 12

ee a | Cc Ge I 1 I v-C I cl A ee yi Se
oo ee a ve. ot Lv v-C S-¢ S-€ So v-C rE ee
4. . 2 ad Co ee 3 aE v-C ci SI ee ro we Ss IA

I I I va ot vl eo c71 I €-T VC v1 yy 2G
oe ee ee re. Te S-€ ve 9-7 €-7 v-C eC | os Glee Jo
oe ee 6-9 ii: 69 Ege S-€ 6-7 8-S 6-S TI-S Sr le es

Sadidpa nuojs sndjod vpi4ioy xOdaf Suaasauvda nlapjnsui vapwsdd siuuad sojodsip siuulfuod 1NpévmoYy DvIIII
-uyol -18uoj] -1U31S
(vWosoulyjuDs ) (DIUDY GDA ) (v1oydo10sg)

("yU09) [ aq

Fig.1.
Fig.2.
Fig.3.
Fig.4.
Fig.5.
Fig.6.
Fig.7.
Fig.8.
Fig.9.
Fig.10.
Fig.11.

Barr and Barr: Psorophora Pupae 15

FIGURES

Psorophora (P.) howardii; metanotum and abdomen.

Psorophora (G.) confinnis; metanotum and abdomen.

Psorophora (P.) ciliata; pigmentation pattern of metanotum and abdomen.
Psorophora (P.) howardii; pigmentation pattern of metanotum and abdomen.
Psorophora (J.) cyanescens; metanotum and abdomen.

Psorophora (J.) longipalpus; metanotum and abdomen.

Psorophora (J.) cyanescens; cephalothoracic setae.

Psorophora (G.) discolor; paddle.

Psorophora (J.) longipalpus; posterior corner of segment IV.

Psorophora (G.) insularia; setae of segment I.

Psorophora (G.) pygmaea; setae of segment I.

FP econtinnis

F elliata

Vill

P howard

a

e ue ae ae

P longipalpus

Wy y

P discolor

: —— (
oy P ecyanescens

ri po

NS AMIR als aT EARMag TE

P insularia

5 wil 1,
RS ey W//

= —;

Zz
ree Zi

ZZ 4
Zi fT»

20 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

INDEX TO SCIENTIFIC NAMES

Aedes 2, 3k, 8

atropalpus, 2

ciliata, 4k, 5, 12-14t; fig.3
columbiae, 8 :

confinnis, 5k, 8, 12-14t; fig.2
Culiseta, 2, 3k, 8

cyanescens, 3k, 4k, 6, 12-14t; fig.5,7
discolor, 4k, 5k, 9, 12-14t; fig.8
ferox, 4k, 6, 7, 12-14t

Grabhamia, 2, 3k, 4k, 6, 8,9, 10, 12-14t
horrida, 3k, 4k, 7, 12-14t

howardii, 4k, 5, 6, 12-14t; fig.1,4
insularia, 4k, 9, 10, 12-14t; fig.10
Janthinosoma, 3k, 6, 7, 8, 12-14t
johnstonii, 3k, 6, 7, 12-14t
longipalpus, 3k, 7, 12-14t; fig.6,9
melanura, 2

Psorophora, 1, 2, 3k, 4k, 5, 6, 12-14t
pygmaea, 4k, 10, 12-14t; fig.11
signipennis, 5k, 9, 12-14t

varipes, 3k, 6, 7, 8, 12-14t

MOSQUITO STUDIES (Diptera, Culicidae)

XIV. REDESCRIPTION OF TOXORHYNCHITES (T.) GIGANTULUS

FROM THE PHILIPPINES
By

O.G.W. Berlin!

T. gigantulus has been known only by the holotype 6 from Limay, Bataan Prov-
ince, and another 6 from San Andales, Rizal Province, both collected by R.C.
McGregor (USNM). In August 1965 I reared individually 3 larvae and 3 pupae of
this species from material brought to our laboratory from the Philippines by Col.
Kremmers of the 5th Epidemiological Flight (PACAF), U.S. Air Force. It is now
possible to describe all the stages of gigantulus. The terminology used in general fol-
lows Belkin (1962).

Toxorhynchites (Toxorhynchites) gigantulus (Dyar & Shannon)
Figs.1,2

1925 Megarhinus gigantulus Dyar and Shannon, 1925:67-68. TYPE: Holotype 6, Limay, Bataan
Province, Luzon, Philippines, Nov 1924, R.C. McGregor [USNM, 28096].

Toxorhynchites (T.) gigantulus of Stone, Knight and Starcke (1959:63).
Megarhinus gigantulus of Edwards (1932:61); Barraud (1934:22); Bohart (1945 :23-24); Stone and
Knight (1957:200).

FEMALE. Wing: 4.0 mm. Proboscis: 5.0 mm. Forefemur: 3.1 mm. Abdomen:
about 4.0 mm. Moderate in size. Head: Integument uniformly dark. Scaling on ver-
tex metallic to bluish green, appearing whitish along orbital line. Orbital (2 pairs)
and interorbital (1 pair) bristles on dorsal side moderately long. Clypeus dark, with
a fine, whitish pubescence. Labial scales purplish. Palpus apparently 4-segmented,
about 0.2 of proboscis length; scales as on proboscis. Antenna about 0.5 of probos-
cis length; torus dark, with a fine, whitish pubescence; flagellar segment 1 with a few
dark scales dorsally; flagellar whorls with 6 moderately long bristles, progressively
smaller distally. Thorax: Integument uniformly black. Mesonotum covered with me-
tallic green scales. Acrostichal, dorsocentral and prescutellar bristles absent; bristles

1Department of Zoology, University of California, Los Angeles, California 90024.

ae Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

on apn, anterior promontory, antealar and supraalar areas, ppl, upper pra, stp and
upper mep distinct. Scales on apn bluish green; scales on ppn as on mesonotum;
ssp, psp and paratergite bare; ppl with silvery scales; pra, stp and most of mep with
silvery scales. Scutellum with a continuous row of dark marginal bristles; scales as
on mesonotum. Legs: Forefemur predominantly dark anteriorly except for basal
creamy scales, posterior surface mostly creamy with a few dark scales distally; most
of midfemur creamy; hindfemur predominantly creamy on both sides except for
dorsal distal dark patch. Tibiae and tarsi of all legs completely dark. Wing: Vein
scales dark. Haltere: Stem pale, knob with creamy scales. Abdomen: Laterotergite
with median metallic green and lateral creamy scales. Scales on tergites‘metallic
green, to purplish depending on angle of incidence of light. Sternites I-VI completely
creamy; distal sternites with basal creamy and apical dark scales. Distinct lateral tufts
of long bristlelike scales on segments VI-VIII; apicolateral tuft on VI completely taw-
ny; lateral tuft on VII tawny basally and dark distally; tuft of segment VIII dark.

MALE. Wing: 4.0 mm. Proboscis: 5.0 mm. Forefemur: 3.1 mm. Abdomen (in-
cluding genitalia): about 5.0 mm. Generally similar to female in size and coloration.
Palpus subequal to proboscis length; 5-segmented, segments 4 and 5 upturned; scaling
predominantly purplish except for creamy ventral scales on distal 0.5 of segments
2,3 and proximal 0.5 of 4. Antenna about 0.5 of proboscis length; flagellar whorls
with 8-10 moderately long bristles, progressively smaller distad. Legs: Claws of fore-
leg and midleg enlarged, unequal; larger claw with a submedian tooth; hindclaws as
in female. Abdomen: Narrower than in female. Apicolateral tuft on segment VI taw-
ny; lateral tuft on segment VII predominantly tawny with a few dark distal scales;
tuft of segment VIII entirely dark. |
~ MALE GENITALIA (fig.1). Ninth tergite lobe shieldlike, with a shallow emargi-
nation apically and with about 10 to 12 small to moderately long apicolateral bris-
tles; basal part constricted. Sidepiece about 3.5 as long as its greatest width, with
scales ventrally and laterally and bristles on all sides. Clasper (without spiniform)
subequal in length to sidepiece; with a small, distinct apical hair, and 2 rows of small
setae in distal 0.6; spiniform 0.25 of clasper. Basal mesal lobe with a strong apically
curved bristle extending beyond middle of sidepiece and a few smaller subapical bris-
tles of varied length. Aedeagus long, swollen at base and produced distally as a nar-
row process ending in a recurved tooth; without a narrow submedian dorsal bridge.
Paraproct long, slender and strongly sclerotized; cercal setae 3 in number.

PUPA (fig.1). Abdomen: 4.6 mm. Trumpet: 0.7 mm. Index about 3.4-3.7. Pad-
dle: 1.35 mm. Chaetotaxy as figured, hairs moderately to darkly pigmented. All
normal hairs present. Cephalothorax: Moderately pigmented. Middorsal ridge mod-
erate. Hair 1-C strongly developed, single; hairs 2,3-C close together; hairs 6,7-C far
forward of 4,5-C; hairs 8,9-C close together and distinctly caudad of trumpet base.
Trumpet: Index about 3.4-3.7; strongly pigmented; pinna moderate. Metanotum:
Moderately pigmented. Hairs 10,11-C removed from 12-C, moderately close to-
gether; an accessory hair caudad of hair 12-C. Abdomen: Integument strongly pig-
mented, progressively lighter on distal segments. Dorsal sensillum on III and IV only.
Hair 1-I dendritic; hair 1-III,VII single to triple, short, usually not reaching apex of
its tergite; hair 5-III-VI strongly developed, single, extending beyond apex of suc-
ceeding segment; hair 6-V,VI strongly developed, single, extending beyond apex of
succeeding segment; hair 6 on other segments short to moderately long, barely
reaching apex of its tergite; hair 8-VI,VII dorsal, cephalad of 6; hair 9-VIII at caudo-
lateral angle, short to moderately long, single. Terminal Segments: Hair 1-IX small,
indistinct; hair 1-X distinct and usually 3 branched (1-5). Paddle: Relatively broad;

Berlin: Toxorhynchites gigantulus : 23

external buttress more or less distinct; margin with distinct hairlike spicules; apex
slightly emarginate at level of midrib. Male genital lobe extending to 0.6 and female
bare to 0.2 of paddle.

LARVA (fig.2). Head: 1.15 mm. Siphon: 0.6 mm. Anal Saddle: 0.4 mm. Chaeto-
taxy as figured, hairs moderately to darkly pigmented. In life larvae appear purplish.
All normal hairs present. Head: Width roughly subequal to length. Mental plate short
and very broad, teeth irregular and strongly developed. Hair 4-C mesad of 1; hairs
5-8-C in a row; hairs 11-13-C in lateral cephalic groups; hairs 14,15-C, single, re-
moved cephalad. Antenna: Length about 0.25 of head; cylindrical and without spic-
ules; hair 1-A inserted beyond 0.75 from base, branched; hairs 2,3-A basad of 1.
Thorax: Hairs 1,2-M on a circular tubercle; hairs 3,4-M on an irregular tubercle, sep-
arate from hair 6; hairs 5,6,7-M on a larger sclerotized plate. Abdomen: Dorsal sen-
silla distinct only on segments III,IV, absent on V. Hairs 1,3-5-I-VII on a common
plate; hair 2-I-VI mesad of 1, on a distinct circular tubercle on segments II-IV; hairs
1,7,11-VII single, spikelike. Segment VIII: Hairs 2-5-VIII arising from a large sclero-
tized plate; hairs 4,5-VIII single, spikelike. Siphon: Short, without acus. Integument
moderately pigmented, smooth. Hair 1-S single, spikelike and within basal 0.3.
Trachea well developed. Anal Segment: Saddle complete, broader than long. Acus
absent. Caudal margin with short to moderately long spicules. Hair 1-X on saddle
margin, single and spikelike; hairs of dorsal brush (2,3-X) branched; ventral brush
with 7 pairs of single, strongly fringed hairs on a boss. Gills short, globose.

SYSTEMATICS. 7. gigantulus is distinguished from other known Philippine Tox-
orhynchites by the following features: (1) in the adults, by the tawny lateral tufts
on abdominal segments VI and VII, (2) in the pupa, by the presence of an accessory
hair on the metanotum caudad of 12-C; hair 3-I long and single; hair 6-VII short and
double and not reaching the apex of its tergite, (3) in the larva, by the single, spike-
like siphonal hair 1-S. As pointed out by Barraud (1934:22), the closest relative of
gigantulus appears to be minimus Theobald, 1905 of Ceylon, India and Sumatra.
The latter differs in the adults in the mainly black tuft of abdominal segment VII
and in the larva in the double or triple hair 1-S.

BIONOMICS. No data are available as to the source of the collections from which
the rearings of this species were made. The breeding sites could have been treeholes,
bamboo or leaf axils. |

DISTRIBUTION. Apparently endemic to the Philippines, where it has been col-
lected only in Central Luzon. Material examined: 15 specimens; 4 6, 2 9, 6 pupae,
3 larvae; 6 individual rearings (3 larval, 3 pupal). |

PHILIPPINES. Luzon: Laguna Province, unspecified locality, 12 Aug 1965, F.E. Baisas et al
(UCLA 249), 2 Ipd (249-22,37), 1 Ip? (249-44), 2 pd (249-103,104), 1 p? (249-101) [UCLA,
USNM]. Limay, Bataan Province, Nov 1924, R.C. McGregor, 6 holotype (Dyar and Shannon 1925:
67, USNM). San Andales, Rizal Province, Dec 1926, R.C. McGregor, 1 6 (Bohart 1945:24, USNM].

REFERENCES CITED

Barraud, P.J.
1934. Family Culicidae. Tribes Megarhinini and Culicini. London, Taylor and
Francis. 463 p. (Fauna of British India, Diptera, v.5)
Belkin, J.N.
1962. The mosquitoes of the South Pacific. v.1. Berkeley, Univ. Calif. Press.
608 p.

24 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

Bohart, R.M.
1945. A synopsis of the Philippine mosquitoes. U.S. Navy, Navmed 580. 88 p.
Dyar, H.G. and R.C. Shannon
1925. The types of Philippine mosquitoes described by Ludlow and other notes
on the fauna. Insecutor Inscitiae Mens. 13:66-89.
Edwards, F.W.
1932. Diptera fam. Culicidae. Jn P. Wytsman, Genera Insectorum, Brussells, Des-
met-Verteneuil, Fasc. 194. 258 p.
Stone, A. and K.L. Knight
1957. Type specimens of mosquitoes in the United States National Museum: VI,
Miscellaneous genera, addenda and summary. Wash. Acad. Sci., J. 47:196-202.
Stone, A., K.L. Knight and H. Starcke
1959. A synoptic catalog of the mosquitoes of the World (Diptera, Culicidae).
Wash., Entomol. Soc. Amer. (Thomas Say Found. Publication 6). 358 p.

FIGURES

Fig.1. TYoxorhynchites (T.) gigantulus; male genitalia and pupa.
Fig.2. Toxorhynchites (T.) gigantulus; larva.

TOXORHYNCHITES

W)

\
))

TFOXORHYNCHITES
Fig2

gigantulus

UCLA 249-37
Laguna Province
Luzon

Philippines

MOSQUITO STUDIES (Diptera, Culicidae)

XV. A NEW SPECIES OF TREEHOLE BREEDING ANOPHELES

FROM THE SOUTHWESTERN UNITED STATES!

By

Thomas J. Zavortink?

During the past 3 years a large number of specimens of the Anopheles commonly
found breeding in treeholes in southern Arizona has been acquired by the project
““Mosquitoes of Middle America.’’ Examination of this material has shown it to be a
species near, but distinct in all stages from, Anopheles barberi. This species, judithae,
is described and illustrated in the present paper. To facilitate its separation from
barberi, that species is redescribed and illustrated.

Terminology and format of the present paper are those of Belkin (1962, 1968).
Abbreviations used for the references follow the “American Standard for Periodical
Title Abbreviations” (Amer. Stand. Ass., 1964). The descriptions of the male geni-
talia and immature stages of barberi and judithae are restricted to features not shown
in the drawings, such as color pattern and variations, and to characters which serve
to separate the 2 species from each other. Since the adults are not illustrated, de-
scriptions of them are somewhat more extensive and contain characters which should
serve to separate the species from each other, from other treehole breeding Anophe-
les, and from the remaining New World Anopheles. The drawings of the preimaginal
forms of each species show the modal condition for the position, length and branch-
ing of the hairs. Since a mature fourth instar larva of barberi in condition suitable
for drawing was not available during this study, the outline of the larval drawing for
that species was copied from judithae. The pecten teeth counts for each species in-
clude not only the large teeth along the posterior margin of the pecten plate, but
also the smaller apical teeth. The spiculation of the larvae is described as it appears
at 200 magnifications. The sidepiece indices are the ratio of the length of the side-
piece as measured from base to apex on the ventral surface and the width at the
point halfway from the base to the apex. In tabulating the bristles on the anterior
surface of the forecoxa, all well-developed setae, including those associated with the

1Contribution from project “Mosquitoes of Middle America” supported by U.S. Public Health
Service Research Grant AI-04379 and U.S. Army Medical Research and Development Command
Research Contract DA-49-193-MD-2478.

2Department of Zoology, University of California, Los Angeles, California 90024.

28 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

coxa-trcchanter joint, have been counted. From the vast literature on Anopheles
barberi, only a few significant papers dealing with the morphology, ecology and dis-
tribution of the species have been cited.

I wish to thank John N. Belkin for reading the rianusctio? and offering valuable
suggestions. I am indebted to Sandra J. Heinemann, Michael J. Nelson and William
A. Powder for the rearing and preparation of specimens, Sally Dieckmann for pre-
paring the final plates and Sheila Bernstein for typing the manuscript for offset re-
production. The new species is dedicated to a special friend, Judith A. Bergland.

Anopheles (Anopheles) judithae Zavortink, n.sp.
Figs. 1,2

TYPES: Holotype 3 with associated larval and pupal skins (UCLA 302-40), Cochise Stronghold
Recreation Area, Dragoon Mountains, Cochise County, Arizona, elevation about 5500 ft, larva
from a rot hole in a living evergreen oak tree, 22 Mar 1966, T.J. Zavortink [USNM]. Allotype ?
with associated larval and pupal skins (UCLA 302-42), same data as holotype. Paratypes: 6 lpd
(302-25 ,41,43,45 46,48), 5 Ip? (302-44,47,49,51,52), 3 pd (302-100,101,103), 1 p? (302-102),
1°, 1P,6L, same data as holotype; 1 Ip? (306-20), same data as holotype except collected on
23 Mar 1966 (UCLA 306); 1 Ipd (328-22), same data as holotype except collected on 4 Sept 1966
(UCLA 328); 7 Ipd (342-13,15,30,33,38,43,44), 11 Ip? (342-10,31,32,34-37,39-42), 10 pd (342-
100-106,109,110,113), 5 p? (342-107,108,111,112,114), 17 6, 25 9,48 P, 27 L, same data as
holotype except collected on 6 Sept 1966 (UCLA 342).

Anopheles barberi of Vargas (1940:319-322; 1943:64,65 66,67, in part); Jenkins and Carpenter
(1946:35-36, in part); Carpenter and LaCasse (1955 :32-34, in part); Richards, Nielsen and Rees
(1956:14); Vargas and Martinez (1956:111-113,140, in part); Rigby, Blakeslee and Forehand
(1963:50); Burger (1965:396); Nielsen, Arnell and Linam (1967:76); Carpenter (1968:72, in
part); Nielsen, Linam, Arnell and Zavortink (1968:363).

FEMALE. Wing: 3.31 mm. Proboscis: 1.92 mm. Forefemur: 2.18 mm. Abdomen:
about 2.1 mm. Head: Integument light to dark brown; orbital bristles dark brown,
interocular bristles amber; dorsum of head with erect scales only, these entirely pale
white or a few lateral ones dark; interocular scales numerous, white, usually none
appreciably elongate; labium and palpus entirely dark scaled, the proximal scales
appressed; torus and first flagellar segment with a few small dark scales. Thorax:
Mesonotum conspicuously shortened and arched; mesonotum with integument very
light to dark brown, largely shining, without a broad hoary median longitudinal
stripe; pleural integument lighter or becoming lighter ventrally; mesonotal bristles
numerous, strongly developed, long and conspicuous, dark in color except for some
amber anterior acrostichals; mesonotal scaling restricted to a small tuft of 5-20 whit-
ish scales in center of anterior promontory and a few erect light or dark scales in
prescutellar area; apn, sp, pra, ppl, stp and upper mep bristles or hairs present, those
on ppl 2-5 in number, those on stp in a more or less continuous vertical row; pleuron
without scales. Legs: Integument of coxae same color as or lighter than adjacent
portions of pleuron; anterior portion of forecoxa with 6-18 strongly developed bris-
tles; coxae without scales; femora, tibiae and tarsi entirely dark scaled. Wing: Veins
and fringe entirely dark scaled, scales on veins uniformly distributed and not grouped
into dark spots. Haltere: Integument of stem light, knob dark scaled. Abdomen: In-
tegument of tergites light to dark brown, usually somewhat dappled, sternites lighter;
completely devoid of scales.

MALE. As for female except for usual sexual differences.

Zavortink: A New North American Anopheles Pores

MALE GENITALIA (fig.1). Sidepiece: More or less cylindrical or short conical,
mesal edge nearly straight, length about 1.9-2.6 width; parabasal spines only slightly
more strongly developed than internal spine; lateral parabasal spine not flattened
preapically, apex gradually attenuate and strongly recurved; internal spine more
strongly developed than large setae of sidepiece, apex usually recurved; distance from
parabasal spines to internal spine usually 1.5-2.0 distance from latter to apex of side-
piece. Clasper: Setae present in basal 0.33. Proctiger: Generally extending to about
level of internal spine.

PUPA (fig.1). Abdomen: 2.80 mm. Trumpet: 0.37 mm. Paddle: 0.83 mm. Cepha-
lothorax: Lightly pigmented with middorsal area, region caudad of trumpet and up-
per portion of wing case darker. Trumpet: Light amber in color. Abdomen: Lightly
pigmented with anterior and/or lateral portions of segments II-VIII darker; hair 9-II-
VIII lightly pigmented, concolorous with integument; 9-III usually about 2.0 diam-
eter and length of 9-II and 0.5 diameter and length of 9-IV; 5-III-VI fine, usually
2,3b; 5-VII usually fine and 2,3b, sometimes thickened and single, but then longer
than 9-VII. Paddle: Lightly pigmented except for darker midrib.

FOURTH INSTAR LARVA (fig.2). Head: 0.67 mm. Anal Saddle: 0.30 mm. Head:
Largely tan to brown in color with subantennal area and collar darker and area of
imaginal eye lighter; mental plate brown, with 3 or 4 teeth on each side of median
tooth; inner clypeals (2-C) closely approximated, separated by a distance less than
that between inner and outer clypeals (3-C); outer clypeals usually double (single,
double); hair 6-C an elongate seta subequal in length to 4-C, usually located nearly
in line with 5,7-C; 14C multiple, 3-6b (2-8). Antenna: Uniformly tan to brown in
color or becoming lighter apically; hair 4-A usually 2,3f. Thorax: Integument with-
out spicules; plumose hairs strongly developed, with lateral branches long and strong-
ly pigmented; hair 2-P with only a few (2-6) long lateral branches; 9-P,M,T simple;
3,4-T usually single (single, double). Abdomen: Integument without spicules; plu-
mose hairs strongly developed, with lateral branches long and strongly pigmented;
hair 1-II-VII similarly developed, palmate; 13-II-V,VII usually single (1-3b); 7-VII
usually 3b (1-3). Segment VII: Hair 5 usually single (single, double). Spiracular
Lobe: Pecten plate with 17-19 (15-20) teeth. Anal Segment: Spicules on posterior
portion of saddle inconspicuous and arranged into short imbricate rows; ventral
brush with 15 or 16 (14-17) hairs, the most caudal of which is normally longer than
2.0 length of saddle.

SYSTEMATICS. Vargas (1940:319-322) was the first to note the occurrence of
a treehole Anopheles in North America west of the continental divide. His record
was based on 3 males collected at Imuris, Sonora, Mexico, a site less than 40 air miles
south of the Arizona border. This material was identified as barberi; although I have
not seen the specimens upon which the report was based, I am certain they are con-
specific with judithae. Richards, Nielsen and Rees (1956:14) first recorded judithae,
as barberi, from Arizona. Their report was based on data supplied by W.A. McDon- |
ald, who had studied only 2 young instar larvae which he had collected near Pata-
gonia. Since judithae is most conspicuously differentiated from barberi in the fourth
instar larva, the failure of earlier workers to detect its specific distinctness was prob-
ably due in large part to the fact that they did not have this stage available for study.

While judithae and barberi form a pair of closely related vicarious species, they are
easily differentiated in all stages. Adults of judithae have numerous white interocular
scales, the erect scales of the head largely pale white, the anterior acrostichal bristles
amber and accompanied by pale scales, and relatively few bristles on ppl and the an-
terior surface of the forecoxa. Those of barberi have the interocular scales reduced

30 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

or absent, the pale erect scales of the head largely dingy yellow, the anterior acros-
tichal bristles dark and usually not accompanied by scales, and relatively numerous
bristles on ppl and the anterior surface of the forecoxa. The male genitalia of the 2
species differ most conspicuously in development of the parabasal spines. In judithae
these spines are only slightly stouter than the internal spine and the lateral parabasal
spine is not flattened preapically. In barberi the parabasal spines are 2-3 times as
stout as the internal spine and the lateral one is flattened preapically. The sidepieces
also differ in shape; the ramifications of this difference, such as the straightness of
the mesal margin, the position of the internal spine relative to the parabasal spines
and the apex of the sidepiece, and the level to which the proctiger projects, are indi-
cated in the descriptions. Pupae of the 2 species are most readily separated on the
basis of hairs 9-III-VIII and 5-VI,VII. In judithae 9-II-VIII is concolorous with the in-
tegument and 9-III is more or less intermediate in length and stoutness between 9-II
and 9-IV, whereas in barberi 9-II alone is concolorous with the integument, 9-III-VIII
being darkly pigmented, and 9-III is more similar to 9-IV than 9-II in length and
stoutness. Hair 5-VI and usually also 5-VII of judithae is fine and divided into
branches, while 5-VI,VII of barberi is thickened, single and shorter than hair 9 of
the corresponding segment. A few specimens from nearly every collection of judithae
have hair 5-VII thickened and single, but when it is developed in this manner it is
longer than 9-VII. The fourth instar larvae of judithae and barberi differ in a striking
number of characters, only the most conspicuous of which are enumerated here.
Larvae of judithae have the inner clypeals separated by a distance less than that be-
tween the inner and outer clypeals, integument of the thorax and abdomen without
spicules, strongly developed plumose hairs, hairs 13-II-V,VII and 5-VIII usually sin-
gle, and spicules of the anal saddle inconspicuous and arranged in imbricate rows.
Larvae of barberi, on the other hand, have the inner clypeals separated by a distance
more than twice that between the inner and outer clypeals, integument of parts of
the thorax and abdomen spiculate, weakly developed plumose hairs, hairs 13-II-V,
VII and 5-VIII usually triple, and spicules of the anal saddle conspicuous and largely
arranged singly.

Anopheles barberi and judithae are apparently quite removed from the other tree-
hole Anopheles placed in the Coelodiazesis group. Adults of these 2 species differ
from barianensis, fausti, plumbeus and xelajuensis in the following: (1) interocular
scales not forming a conspicuous white elongate tuft, (2) erect scales of head not
sharply divided into brilliant white ones mesally and very dark ones laterally, (3)
mesonotum conspicuously shortened and arched, giving a Culex-like aspect, (4) mes-
onotum without a conspicuous broad hoary median longitudinal stripe and (5) acros-
tichal scaling of mesonotum much reduced. They differ from all of the above-men-
tioned species except plumbeus by lacking light scales in the wing fringe and/or at
the apices of the hindfemur and hindtibia. It should be pointed out, however, that
two species of trechole breeding Anopheles from the western coast of India, culici-
formis and sintoni, which are not placed in the Coelodiazesis group, agree with bar-
beri and judithae in all of the features mentioned above.

The larvae of the 2 other New World treehole breeding Anopheles differ from
barberi and judithae in a number of characters. Both fausti and xelajuensis have hair
3-C laterad of 4-C; fausti has hair 9-III-VI elongate and plumose and hairs 5,7-VII
weakly developed; xelajuensis has conspicuous stellate body hairs and does not have
palmate hairs. :

Larval hair O-P has not been located on barberi, fausti, judithae or plumbeus and
is apparently absent. The taxonomic significance, if any, of the presence or absence
of this hair is unknown at the present time.

Zavortink: A New North American Anopheles 31

BIONOMICS. The bionomics of judithae are summarized in Nielsen, Arnell and
Linam (1967:76) and Nielsen, Linam, Arnell and Zavortink (1968:363). |

Larvae of this species, like barberi, are more frequently found in permanent than
temporary treeholes. In southern Arizona the species is common in the high eleva-
tion xeric evergreen forest (Arbutus, Cupressus, Juniperus, Pinus and Quercus), the

‘moderate elevation riparian woodland (Fraxinus, Platanus, Populus, Salix and Sam-
bucus) and the low elevation gallery forest (Populus and/or Salix).

DISTRIBUTION. Ajn. judithae has been found in Arizona, southwestern New Mex-
ico (Nielsen, Linam, Arnell and Zavortink 1968:363) and Sonora (Vargas 1940:319-
322). During this study only the Arizonan specimens cited below have been exam-
ined. All this material is in the UCLA collection. Material examined: 810 specimens;
154 6, 149 9, 279 pupae, 228 larvae; 147 individual rearings (92 larval, 47 pupal, 8
incomplete).

UNITED STATES. Arizona: Baboquivari Mountains, Kitt Peak National Observatory, 12 Sept
1968, T.J. Zavortink (UCLA 448), 1 lpd (448-27), 6 lp? (448-20-25), 1 p? (448-15), 1 lp (448-28).
Dragoon Mountains, Cochise Stronghold Recreation Area, type series, see above. Nogales (13 mi
NW), Calabasas Picnic Ground, 21 Mar 1966, T.J. Zavortink (UCLA 297), 2 Ip? (297-20,21). No-
gales (9 mi NE), 21 Mar 1966, T.J. Zavortink (UCLA 298), 3 lpd (298-40,42,49), 4 lp? (298-43,
45,50,51), 3 pd (298-101-103), 4 Ip (298-41,46-48), 3 2, 4 P, 22 L; same data (UCLA 299), 5 Ipé
(299-21 ,26,28-30), 6 Ip? (299-20,23-25,27,31), 1 lp (299-22), 2 L. Patagonia (2-4 mi WSW),
Sonoita Creek, 24 Aug 1954, W.A. McDonald (UCLA 137), 2 L; 6 Sept 1963, J. Burger (UCLA
413A), 2 L; 18 Aug 1964, J. Burger (UCLA 253), 17 6, 11 9; 13 Sept 1964, J. Burger (UCLA
270), 3 L; 20 Sept 1964, J. Burger (UCLA 260), 17 6, 7 9; 25 July 1965, J. Burger (UCLA 281),
4 L; 27 July 1965, J. Burger (UCLA 282), 5 L; 21 Mar 1966, T.J. Zavortink (UCLA 300), 2 Ipé
(300-30,38), 6 lp? (300-32-36,39), 2 pd (300-100,102), 2 p? (300-101,103), 3 6,3 2,6 P, 12 L;
5 Sept 1966, T.J. Zavortink (UCLA 333), 1 6, 1 P, 2 L; same data (UCLA 334), 1 Ipd (334-15),
4 Ip? (334-12-14,16), 2 pd (334-102,103), 3 p? (334-100,101,104), 1 9, 3 P, 2 L; 14 Sept 1968,
T.J. Zavortink (UCLA 458), 1 p? (458-100). Portal (15 mi WNW), 6 Sept 1966, T.J. Zavortink
(UCLA 343), 9 Ipd (343-14,15,17,31-34,36,37), 8 Ip? (343-12,16,19,30,38-41), 10 pd (343-100-
104,107,109,110,113,114), 4 p? (343-105 108,111,112), 2 lp (343-13,35), 30 6, 25 °, 69 P, 39 L.
Prescott (7 mi NE), 1 Sept 1966, T.J. Zavortink (UCLA 315), 2 Ipd (315-16,18). Santa Rita
Mountains, 20 Oct 1940, R.A. Flock, 1 d.

Anopheles (Anopheles) barberi Coquillett
Figs.3 ,4

1903.Anopheles barberi Coquillett, 1903:310. TYPE: Holotype 9, Plummer’s Island, Maryland,
17 Aug 1903, H.S. Barber (USNM, 6959).

Anopheles barberi of Dyar (1904:243-244; 1928:454); Vargas (1942:329-331; 1943:64,65 66,67,
in part); Matheson (1944:114-115); Jenkins and Carpenter (1946:35-36, in part); Darsie (1949:
524-525); Penn (1949:65-66); Carpenter and LaCasse (1955:32-34, in part); Vargas and Mar-
tinez (1956:111-113,140, in part); Carpenter (1968:72, in part).

Coelodiazesis barberi of Howard, Dyar and Knab (1917:1036-1038).

FEMALE. Wing: 3.49 mm. Proboscis: 1.73 mm. Forefemur: 2.10 mm. Abdomen:
about 2.2 mm. Very similar to judithae but differing in the following features. Head:
Interocular bristles dark or amber; erect scales on dorsum of head dingy yellowish
mesally, dark or dingy yellowish laterally; interocular scales absent or very few. Tho-
rax: Anterior acrostichal bristles dark in color; center of anterior promontory usu-
ally without scales, sometimes with 1-4 narrow dark scales; ppl bristles 7-11. Legs:
Anterior forecoxal bristles 20-28.

32 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

MALE. As for female except for usual sexual differences.

MALE GENITALIA (fig.3). Sidepiece: More or less elliptical, mesal edge usually
distinctly convex, length about 1.7-2.1 width; parabasal spines much more strongly
developed than internal spine; lateral parabasal spine flattened preapically, apex
sharply attenuate and recurved or sinuous; internal spine not more strongly devel-
oped than large setae of sidepiece, apex usually sinuous; distance from parabasal
spines to internal spine about 0.8-1.0 distance from latter to apex of sidepiece.
Clasper: Usually without setae in basal 0.33. Proctiger: Usually extending to near
level of apex of sidepiece.

PUPA (fig.3). Abdomen: 3.04 mm. Trumpet: 0.37 mm. Paddle: 0.88 mm. Cepha-
lothorax: Moderately pigmented, with dark areas a little more extensive than in
judithae. Trumpet: Deep amber in color. Abdomen: Moderately pigmented, with
dark areas somewhat more extensive than in judithae; hair 9-II moderately pigmen-
ted, concolorous with integument, 9-III-VIII darkly pigmented, strongly contrasting
with integument; 9-III usually approaching 9-IV in both diameter and length; 5-III-
V fine, usually single; 5-VI,VII thickened and single, shorter than hair 9 of corre-
sponding segment. Paddle: Moderately pigmented except for darker midrib.

FOURTH INSTAR LARVA (fig.4). Head: 0.65 mm. Anal Saddle: 0.27 mm. Head:
Largely brown in color with pattern of pigmentation as in judithae; mental plate
dark brown, usually with 5 or 6 (4-6) teeth on each side of median tooth; inner cly-
peals (2-C) widely spaced, separated by a distance greater than 2.0 that between in-
ner and outer clypeals (3-C); outer clypeals usually single (single, double); hair 6-C
a short pointed peg only about 0.1 length of 4-C, located far caudad of 5-C; 14-C
single (single, double). Antenna: Uniformly brown to dark brown in color or with
apex lighter; hair 4-A single. Thorax: Integument, especially on underside of pro-
thorax, spiculose; plumose hairs weakly developed, with lateral branches short and
often lightly pigmented; hair 2-P with numerous (more than 8) short lateral branches;
9-P,M,T spiculate; 3,4-T usually 2,3b (1-4). Abdomen: Integument on underside of
segments I-VIII spiculose; plumose hairs weakly developed, with lateral branches
short and often lightly pigmented; hair 1-IT usually not palmate, 1-4b, conspicuously
less well developed than 1-III-VII when palmate; 1-III-VII palmate, similarly devel-
oped or 1-VII smaller; 13-IJ-V,VII normally 3b (3-5); 7-VII usually single (single,
double). Segment VII: Hair 5 usually 3b (2,3). Spiracular Lobe: Pecten plate with
12-15 (11-17) teeth. Anal Segment: Spicules on posterior portion of saddle more
conspicuous than in judithae and, for the most part, not arranged into rows; ventral
brush with 17 or 18 hairs, the most caudal of which is normally only 1.0-1.5 length
of saddle.

SYSTEMATICS. The diagnostic features of this species are indicated above in the
description and are discussed in the systematics section for judithae. The small series
of larvae of this species which has been studied is remarkably uniform except for
the development of hair 1-II. This hair varies from a weakly developed simple seta
to a well-developed palmate hair. Material from Ohio exhibits this entire range of
variation within one collection. While hair 1-VII is palmate in all specimens exam-
ined during this study, it may not always be developed in that manner (Howard,
Dyar and Knab 1917:1036-1038; Dyar 1928:454).

BIONOMICS. The bionomics of this species are summarized in jest and Car-
penter (1946:35-36) and Carpenter and LaCasse (1955 :32-34).

The immature stages are more frequently found in permanent treeholes in associ-
ation with species of Orthopodomyia and Toxorhynchites than in temporary tree-
holes along with species of Aedes. Two collections made in Ohio in December of

Zavortink: A New North American Anopheles 33

1967 suggest that the 2 species of Orthopodomyia inhabiting the eastern United
States, alba and signifera, may select different types of permanent treeholes for ovi-
position and that barberi selects the same type as signifera. The collections referred
to were from horsechestnut trees (Aesculus hippocastanum) growing within 40 me-
ters of each other; the relatively shallow, brown-colored water of a hole with a broad
opening contained small numbers of alba, barberi and signifera larvae, while the deep,
yellow-colored water of a hole with a narrow opening contained several hundred alba
larvae and a single signifera larva.

DISTRIBUTION. Anopheles barberi has been found throughout the eastern Uni-
ted States, from South Dakota and New York in the north to Texas and Florida in
the south (Carpenter and LaCasse 1955:32-34; Carpenter 1968:72). During this
study only the material cited below, all of which is in the UCLA collection, has
been seen. Material examined: 75 specimens; 17 6, 15 9, 23 pupae, 20 larvae; 20
individual rearings (15 larval, 3 pupal, 2 incomplete).

UNITED STATES. Alabama: Guntersville Lake, 28 May-27 Aug 1942, 3 . Sheffield, 18 Aug
1942, J.N. Belkin, 1 2. Wilson Dam, 10 June-19 July 1942, J.N. Belkin, 1 lp? (167), 1 pd (171),
1 p? (168), 2 6. Arkansas: Little Rock, Little Fourche Bayou, 28-29 Mar 1943, J.N. Belkin, 2 lp
(440,448), 3 P. Maryland: Baltimore, Patapsco State Park, 17 Oct 1965, W.A. McDonald (UCLA
286), 1 L. Cabin John, July 1965, W.A. McDonald (UCLA 286A), 1 L. New York: Ithaca, 18-26
July 1932, R. Matheson, 2 6,3 2. Ohio: Canton, 22 Dec 1967, T.J. Zavortink (UCLA 437), 10 lpd
(437-30-36,39,41,43), 4 lp? (437-37,38,40,42), 1 pd (437-44), 1 L. Tennessee: Kentucky Lake,
26 Aug 1942, 1 6. Texas: Abilene, 30 Aug 1953, R.X. Schick, 2 9.

REFERENCES CITED

Belkin, John N.
1962. The mosquitoes of the South Pacific (Diptera, Culicidae). v.1. Berkeley,
Univ. Calif. Press. 608 p.
1968. Mosquito Studies (Diptera, Culicidae). VII. The Culicidae of New Zealand.
Amer. Entomol. Inst., Contrib. 3(1). 182 p.
Burger, John F.
1965. Aedes kompi Vargas and Downs 1950, new to fhe United States. Mosquito
News 25 :396-398.
Carpenter, Stanley J.
1968. Review of recent literature on mosquitoes of North America. Calif. Vector
Views 15:71-98.
Carpenter, Stanley J. and W.J. LaCasse ©
1955. Mosquitoes of North America (North of Mexico). Berkeley, Univ. Calif.
Press. 360 p.
Coquillett, Daniel W.
1903. A new Anopheles with unspotted wings. Can. Entomol. 35:310.
Darsie, Richard F.
1949. Pupae of the anopheline mosquitoes of the northeastern United States (Dip-
tera, Culicidae). Rev. Entomol. 20:509-530.
Dyar, Harrison G.
1904. Brief notes on mosquito larvae. N.Y. Entomol. Soc., J. 12:243-246.
1928. The mosquitoes of the Americas. Wash., Carnegie Inst. [Publication 387]
616 p.
Howard, Leland O., H.G. Dyar and F. Knab
1917. The mosquitoes of North and Central America and the West Indies. v.4.
Wash., Carnegie Inst. [Publication 159] p.525-1064.

34 Contrib. Amer. Ent. Inst., vol. 4, no. 4, 1969

Jenkins, Dale W. and S.J. Carpenter

1946. Ecology of the tree hole breeding mosquitoes of Nearctic North America.

Ecol. Monogr. 16:31-47.
Matheson, Robert

1944. Handbook of the mosquitoes of North America. ed.2. Ithaca, Comstock
Publishing Co. 314 p.

Nielsen, Lewis T., J H. Arnell and J.H. Linam

1967. A report on the distribution and biology of the tree hole mosquitoes in the
western United States. Calif. Mosquito Control Ass., Proc. Pap. 35:72-76.

Nielsen, Lewis T., J.H. Linam, J H. Arnell and T.J. Zavortink

1968. Distributional and biological notes on the tree hole mosquitoes of the west-

ern United States. Mosquito News 28:361-365.
Penn, George H.

1949. Pupae of the Nearctic anopheline mosquitoes north of Mexico. Nat. Malaria
Soc., J. 8:50-69.

Richards, Charles S., L.T. Nielsen and D.M. Rees

1956. Mosquito records from the Great Basin and the cocaine of the lower Colo-
rado River. Mosquito News 16:10-17.

Rigby, Paul T., T.E. Blakeslee and C.E. Forehand

1963. The occuirence of Aedes taeniorhynchus (Wiedemann), Anopheles barberi

(Coquillett), and Culex thriambus (Dyar) in Arizona. Mosquito News 23:50.
Vargas, Luis

1940. Anopheles (Anopheles) barberi en Mexico. Rev. Inst. Salubr. Enferm. Trop.
1:319-322.

1942. El huevo de Anopheles barberi Coquillett, 1903. Rev. Inst. Salubr. Enferm.
Trop. 3:329-331.

1943. Los subgeneros Americanos de Anopheles (Diptera, Culicidae). Anopheles
(Russellia) xelajuensis de Leon, 1938 n.subgn. y Anopheles (Coelodiazesis)
fausti n.sp. Rev. Inst. Salubr. Enferm. Trop. 4:57-77.

Vargas, Luis and Amado Martinez Palacios

1956. Anofelinos Mexicanos. Taxonomia y distribucion. Mexico, D.F., Secretar.

Salubr. Asistencia. 181 p.

FIGURES

1. Anopheles (An.) judithae; male genitalia and pupa
2. Anopheles (An.) judithae; larva

3. Anopheles (An.) barberi; male genitalia and pupa
4. Anopheles (An.) barberi; larva

| Oat 8S AOr ri

780

ANOPHELES

judithae
Ui CLA 302

ge
Le \

barberi
UCLA 437

United States

~ANOPHELES

io
United States

DENS
accel
De Na ak

ce

Poke Ae a a
SSS ath
SRT

Leone!
Tite V3
ty

wee
ayy

ar Ne Hoban *

a,

BE Tra)

‘ ea
anes

No.

No.

(Continued from inside front cover)
Parts of Volume 4, with prices

Mosquito studies (Diptera, Culicidae), 1968.
X. Peters, T. Michael. Dixinae originally described from North America. 7 pages.
XI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann. Mosquitoes
originally described from Argentina, Bolivia, Chile, Paraguay, Peru, and Uruguay.
29 pages. Price: $1.25, postpaid.

Mosquito studies (Diptera, Culicidae), 1969.
XII. Berlin, O. G. W. A revision of the Neotropical subgenus Howardina of Aedes.

Carlson, Robert W. and Fred B. Knight, 1969.
Biology, taxonomy, and evolution of four sympatric Agrilus beetles (Coleoptera:
Buprestidae). 104 pages, 47 figures. Price: $3.25, postpaid.

Mosquito studies (Diptera, Culicidae), 1969.
XIII. Barr, Ralph A. and Sylvia Barr. Pupae of the genus Psorophora in America
north of Mexico. 19 pages, 11 figures.
XIV. Berlin, O. G. W. Redescription of Toxorhynchites (T.) gigantulus from the
Philippines. 7 pages, 2 figures.
XV. Zavortink, Thomas J. A new species of treehole breeding Anopheles from
the southwestern United States. 12 pages, 4 figures. Price for XIII, XIV, and XV:
$1.25, postpaid.

Se alan

ae =
ag

ae

By

Sie
aay

dL.

1S /
DS /
Wwe
Contributions
of the

American Entomological Institute

Volume 4, Number 5, 1970

A REVISION OF THE GENUS SPATHIUS IN AMERICA
NORTH OF MEXICO (HYMENOPTERA, BRACONIDAE)

by
Robert W. Matthews

ra HSON; Ay

MAY 1 8 1970
CiBRARIE2

CONTRIBUTIONS
of the
AMERICAN ENTOMOLOGICAL INSTITUTE

The Contributions are for larger papers on insects. Each paper is a separate number, with

separate pagination and index. Separate numbers aggregating about 500 pages constitute a volume.
Issues appear irregularly, as suitable manuscripts are available.

Copies are sold separately or in subscriptions to complete volumes. Complete volumes are

$12.00. The price of separate numbers varies. Subscribers are billed for each volume with its
beginning number, and receive the parts as issued. Orders for separate numbers that total less
than $8.00 must be accompanied by payment. Address orders or correspondence to the American
Entomological Institute, 5950 Warren Road, Ann Arbor, Michigan 48105, U.S.A.

No.
No.

No.

No.

Parts of Volume 1, with prices

Dasch, Clement E., 1964. The Neotropic Diplazontinae (Hymenoptera, Ichneumonidae).
75 pages, 69 figures. Price: $2.25, postpaid.

Mosquito studies (Diptera, Culicidae), 1965.
I. Belkin, John N., et al. A project for a systematic study of the mosquitoes of
Middle America. =
II. Belkin, John N., et al. Methods for the collection, rearing and preservation of
mosquitoes. aes
78 pages, 4 figures. Price: $2.25, postpaid.

. Same as no. 2, but in Spanish. Price: $2.25, postpaid.

Matthews, Robert W., 1965. The biology of Heriades carinata Cresson. (Hymenoptera,
Megachilidae). 33 pages, 23 figures. Price: $1.00, postpaid.
Mosquito studies (Diptera, Culicidae), 1965. .
III. Ramalingam, Shivaji, and John N. Belkin. Two new Aedes from Tonga and Samoa.
IV. Belkin, John N. The mosquitoes of the Robinson-Peabody Museum of Salem
expedition to the southwest Pacific, 1956.
34 pages, 3 figures. Price: $1.00, postpaid.
Mosquito studies (Diptera, Culicidae), 1965.
V. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.
Mosquitoes originally described from Middle America. 95 pages. Price: $2.75 postpaid.
Mosquito studies (Diptera, Culicidae), 1966.
VI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann.
Mosquitoes originally described from North America. 39 pages. Price: $1.00, postpaid.
DeLong, Dwight M. and Paul H. Freytag, 1967. Studies of the world Gyponinae
(Homopotera, Cicadellidae). A synopsis of the genus Ponana.
86 pages, 257 figures. Price: $2.50, postpaid.
Contributions to the mosquito fauna of southeast Asia, 1967.
I. Delfinado, Mercedes D. The genus Aedes, subgenus Neomacleaya Theobald in
Thailand. 55 pages, 20 figures. Price: $1.75, postpaid.

Parts of Volume 2, with prices

Contributions to the mosquito fauna of southeast Asia, 1967.
II. Bram, Ralph A. The genus Culex in Thailand (Diptera: Cuneta),
296 pages, 104 figures. Price: $9.00, postpaid.
Schuh, Toby, 1967. The shore bugs (Hemiptera: Saldidae) of the Great Lakes region.
35 pages, 30 figures. Price: $1.25, postpaid.
Snider, Richard J., 1967. The chaetotaxy of North American Lepidocyrtus s. str.
(Collembola: Entomobryidae). 28 pages, 57 figures. Price: $1.00, postpaid.
Contributions to the mosquito fauna of southeast Asia, 1968.
WI. Delfinado, Mercedes D. The genus Aedes, subgenus Neomacleaya Theobald
of southeast Asia. 76 pages, 26 figures. Price: $2.25, postpaid.
Contributions to the mosquito fauna of southeast Asia, 1968.
IV. Knight, Kenneth L. Species of the subgroup Chrysolineatus of group D, genus
Aedes, subgenus Finlaya Theobald. 45 pages, 12 figures. Price: $1.50, postpaid.

Parts of Volume 3, with prices

Mosquito studies (Diptera, Culicidae), 1968.
VII. Belkin, John N. The Culicidae of New Zealand.
182 pages, 30 figures. Price: $5.50, postpaid.
Mosquito studies (Diptera, Culicidae), 1968.
VIll. Zavortink, Thomas J. A prodrome of the genus Orthopodomyia.
221 pages, 41 figures. Price: $6.75, postpaid.
Fleming, Richard C., 1968. Head musculature of sphinx moths (Lepidoptera:
Sphingidae). 32 pages, 37figures. Price: $1.00, postpaid.
Mosquito studies (Diptera, Culicidae), 1968.
IX. Belkin, John N. The type specimens of New World mosquitoes in European
museums. 69 pages. Price: $2.00, postpaid.

(Continued on back cover)

A REVISION OF THE GENUS SPATHIUS IN AMERICA
NORTH OF MEXICO (HYMENOPTERA, BRACONIDAE)*

By

Robert Wendell Matthews**

CONTENTS
DS ON a ee ee eae 6 ae Co Ae ee 3
COLLECTIONS AND ACKNOWLEDGEMENTS . 2 6 6.006 6s 1 ee ete 3
TERMINOLOGY AND SPECIFIC CHARACTERS ...°...00 6 00 eee es © 4
Dae GENUS Oe UI, eo ee ee a ee eee ea ee 10
EVORUTIONARY THE N6e fi ee i a a 12
BOM GN 6 6 igs ar eis oo A We keV gp as ee eg ese 16
KEY TO THE NORTH AMERICAN SPECIES OF SPATHIUS........ 20
THE NORTH AMERICAN SPECIES OF SPATHIUS
Bs OPAChy Wels AGHNICNG, 6 66 ee ee ee 26
De DECyipatiue, SPL Ne a ele a ow ye oe 29
Dr eel, Sy Ne ec OG ee Oa ele) ee wee we 30
a) BUA a eet Eee Ge a kd ee 32
ey er ee 35
By GONeIDelIO late AGNIMCAG to ie a ee a ee 38
Oren eek kak ee ee we We BS ee kk ae 39
ey Vee SO ee eee Go ew ee eee 42
De MCUUOISS BeNMICR 2 a ea ne wees 44
Oe CANAOCNSIS OUINCAG ee er a eo a 8 47
Oe) CRMCNOCS, SU Ne ee ee ae a a 51
Wo De UG ACC ee ee ae A a we 53
Oe OARS, SO 8 ee, ke ee 55
RUS Re ee oa eg a ee i Gs ge 56
DOOM S CE We a ag ee ee a a ee 59
Ry OR MereClOR Ge ee ee a ae ae Ay eg ae “61
DL INNING) PTOVANCNCT ec aa ee ee ek Ve ee ks 64
mo PMO CIOUUS TILCY ee tar ee ae Bh sa a as as 66

*k
Submitted to the Department of Biology, Harvard University, in partial fulfillment
of the requirements for the degree of Doctor of Philosophy.
This study is intended as a contribution to the series of systematic studies of the
subfamily Doryctinae for the Nearctic Region, initiated by Marsh (1965).
**Present address: Department of Entomology, University of Georgia, Athens,
Georgia 30601, U.S.A.

Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

o DPBRNOUS A SHMICAT oe bia ek we Re. ne, ee ee ea
ATEILE LETTS PROTO ie) shoes eck Goll des ek en WE Re ee a
S PPO PICe se ASE ae Wi a ee ee a ae OR ecw a

SICH Tr es ae es RS OO Sa es
SUMMARY SLRS IISr WRU ise Ran TE RNS SERBS aif b Aol Per 4D Sea eRe
PATON AT UR CIPRO oo ew ea Ss Se Oe ae ae owe es

INDEX

INTRODUCTION

Spathius is a cosmopolitan genus of Braconidae with over 300 described
forms which, with few exceptions, are ectoparasites of wood-boring coleop-
terous larvae, predominantly Scolytidae. A relatively recent revision of the
Old World species exists (Nixon, 1943). For the Nearctic species, the only
work to date consists of a few isolated and vague descriptions, and Ashmead’s
(1893) key to the species (including several newly described ones). This has
proved unusable because of several errors and inconsistencies. The 16 de-
scribed Nearctic species (Muesebeck et al., 1951) have been so poorly under-
stood that identification has been impossible except by comparison with the
types.

This revision recognizes 21 species of North American Spathius, of which
11 are described as new. The types of all previously described species have
been examined and new descriptions prepared for all valid species. Of the
more than 2500 specimens seen, a large proportion represent reared series.
In addition to providing host data, these series have permitted positive sex
associations and made it possible to more accurately assess intraspecific
variation.

COLLECTIONS AND ACKNOWLEDGEMENTS

The following institutions and individuals loaned the material included in
this study:

Academy of Natural Sciences of Philadelphia, Philadelphia,
Pennsylvania (ANSP)

American Entomological Institute, Ann Arbor, Michigan (Townes)

American Museum of Natural History, New York, N. Y. (AMNH)

Boston University, Boston, Massachusetts (BU)

California Academy of Sciences, San Francisco, Calif. (CAS)

Canadian National Collection of Insects, Ottawa, Canada (CNC)

Connecticut Agricultural Experiment Station, New Haven, Conn.
(CAES)

Cornell University, Ithaca, New York (CU)

Eastern Illinois University, Charleston, Ill. (Riegel)

Henry Hespenheide, Philadelphia, Pennsylvania (HH)

filinois Natural History Survey, Urbana, Ill. (INHS)

Kansas State University, Manhattan, Kansas (KSU)

Michigan State University, East Lansing, Mich. (MSU)

Museum of Comparative Zoology, Harvard University, Cambridge,
Mass. (MCZ)

North Carolina State University, Raleigh, N.C. (NCS)

Oregon State University, Corvallis, Ore. (OSU)

Charles Porter, Metuchen, New Jersey (Porter)

Syracuse University, Syracuse, New York (Syracuse)

University of Arizona, Tucson, Ariz. (Arizona)

University of California, Berkeley, Calif. (Berkeley)

University of Idaho, Moscow, Idaho (Idaho)

University of Minnesota, St. Paul, Minn. (Minnesota)

University of Montreal, Montreal, Canada (Montreal)

University of Wisconsin Insectarium and Roy D. Shenefelt, Madison,
Wisconsin (RDS)

4 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

U.S. Department of Agriculture, Southern Forestry Experiment
Station, Pineville, Louisiana (Pineville)

U.S. Department of Agriculture, Northeastern Forest Experiment
Station, Hamden, Connecticut (Hamden)

U.S. National Museum, Washington, D.C. (USNM)

I am indebted to the curators and owners of the above collections for mak-
ing these available. I also thank Drs. R. T. Franklin (University of Georgia),
W. B. Becker (University of Massachusetts), and B. H. Kennedy (Northeast-
ern Forest Experiment Station, Delaware, Ohio) for sending living and pre-
served specimens from their reared material. Many others have sent host
records and other data; to allIam most grateful. Additional thanks are due
to Dr. Stephen L. Wood (Brigham Young University) for checking the synonymy
of the Scolytidae reared as hosts of Spathius.

Financial support during the course of this study included: University,
National Defense Education Act, Title IV and Richmond fellowships from Har-
vard University; travel funds from National Science Foundation Grant GB 3167
to the Evolutionary Biology program of Harvard University; the 1967 Summer
Fellow award of the Edmund Niles Huyck Preserve, Inc., Rensselaerville,
New York; and a Smithsonian Institution Research Assistantship Award during
May, 1968.

Among the many individuals offering me aid and encouragement during
this study, special acknowledgement is gratefully given to:

Dr. Howard E. Evans, my graduate advisor, for his unflagging interest,
helpfulness and encouragement during all phases of my stay at Harvard Uni-
versity.

Professors E. O. Wilson and F. M. Carpenter, the other members of my
thesis committee, who have freely given of their time and energy in my behalf.

Dr. Paul M. Marsh of the U. S. National Museum, for his inspiration and
perceptive interest, andfor, during my visit in May, 1968, giving generously
of his time, energy and office space.

Professor Roy D. Shenefelt of the University of Wisconsin, for kindly
thus enabling me to move much more swiftly from bibliographic to taxonomic
considerations.

Finally, to my wife, Janice, for her patience, aid and understanding
throughout my graduate career.

TERMINOLOGY AND SPECIFIC CHARACTERS

The descriptive vocabulary used in the species descriptions is taken largely
from Nixon’s (1943) revision of the Old World Spathiinae and Marsh’s (1965) re-
view of the Nearctic Doryctinae. However, to minimize confusion, diagrams
have been prepared illustrating various body parts and showing the points be-
tween which measurements have been made.

Measurements were made with the aid of an ocular micrometer, the ma-
jority at a magnification of 72x on a Leitz stereoscopic microscope. All
quantitative measurements, with the exception of body and wing length, are
presented in the form of proportions. Where possible, measurements were
taken on at least ten individuals of each sex, selected to represent as wide a
geographic range as possible. Care was taken to be as accurate as possible,
but it is recognized that differences in viewing angle as well as personal idio-
syncrasies of technique will give slightly different results for different workers.

Matthews: Nearctic Spathius 5)

In this regard it is important to adjust the specimen so that the end points of
the linear measurement fall in the same focal plane. Drawings of wing vena-
tion and male genitalia were prepared from material mounted on slides and
projected with a Bio-Scope. All other drawings were made on squared paper
with the aid of a squared ocular disc.

HEAD.- In Spathius, as in most Doryctinae, the head is slightly broader
than long, with the temples broad behind the eyes, giving the head a subcubi-
cal shape. The occiput is always completely margined by the occipital carina;
in the Nearctic species, this carina is never fused with the hypostomal carina,
although such fusion occurs in several species groups of Old World Spathius,
and is usually correlated with a narrowing of the temples. (One Old World
species, Spathius aristaeus Nixon, differs from all other Spathius in the lack
of a completely margined occiput. ) :

The antennae of Spathius are usually slender, filiform, and as long as or
longer than the body; the number of segments is highly variable within a spe-
cies (for example, 25-36 segments in S. trifasciatus), and the right and left
antennae may differ in number of segments on the same individual. In the de-
scriptions, number of flagellomeres is used and refers to number of flagellar
segments only, excluding the scape and pedicel. In the brachyurus group, the
flagellomeres are characteristically shortened and subcubical, each with a
corona of distinctly longer setae at its apex; also the number of flagellar seg-
ments in this group is the lowest for any Spathius. On close examination under
high power, the eyes of most if not all Spathius have minute hairs between the
facets, these being most obvious in members of the brachyurus group.

The maxillary palpus of Spathius is six-segmented and typically somewhat
elongate, reaching nearly to the mesocoxal cavities. However, the brachyurus
group exhibits a tendency toward reduction in palpal size, reaching an extreme
in S. brevipalpus, sp. n. The mandibles are always bidentate, their bases
striate ventrally, and are of little taxonomic value.

Head sculpture*, an important species character, is described with refer-
ence to four major regions as follows (Figs. 1, 2):

Face: area between clypeus and antennal insertions; it is almost always
transversely strigose-granular in Nearctic Spathius and of little tax-
onomic value.

Frons: area between antennal insertions and median ocellus; it has typi-
cally transverse striation of varying intensity, but is rugose or smooth
in some species.

Vertex: dorsal region of head between lateral ocelli and occipital carina.
Presence or absence of sculpture here is an important species character;
when present it usually occurs as transverse striations of varying in-
tensity.

Temples: area between eyes and occipital carina; some Nearctic Spathius,
in contrast to most doryctines, exhibit fairly prominent sculpturing here.

Midwras in this study Eady (1968) published a paper illustrating types of micro-
sculpture in the Hymenoptera and the terminology used to describe them. In general
my interpretations are in agreement, and his paper should be consulted for clarifica-
tion of any descriptive terminology concerning sculpture.

6 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

vertex

temple

“nr,
4

~~
4 wicdoe

0

Figures 1 and 2. Head of Spathius floridanus Ashmead, frontal
and lateral views respectively; explanations of measure-
ments and terminology given in text.

Additional very useful characters are size and position of the eyes and
ocelli; following are measurements and abbreviations for expressing these
features (Figs. 1, 2):

Eye Height (EH): maximum height (or length) of eye measured in lateral

view.

Eye Width (EW): maximum width of eye measured in lateral view.

Malar Space (MS): least distance between lower eye margin and mandible

base.

Ocello-eye Distance (OED): least distance between a lateral ocellus and

eye margin on same side.

Ocello-occiput Distance (OOD): least distance between occipital carina

and a line tangent to posterior margins of lateral ocelli.

Ocellar Triangle Width (OTW): maximum distance between (and including)

lateral ocelli.

Temple Width (TW): least distance between occipital carina and posterior

dorsal margin of eye.

Interocular Distance (IOD): least width of face along a line just below

antennal insertions.

Matthews: Nearctic Spathius 7

Figure 3. Fore wing of S. floridanus, 2. Veins: A, recurrens; B, first abscissa of
cubitus; C, first intercubitus; D, E and F, first, second and third abscissae of
radius; G, second intercubitus; H, subdiscoideus; J, second abscissa of cubitus.
Cells: 1, radial; 2, 3 and 4, cubital; 5 and 6, discoidal; 7, first brachial cell.

Figure 4. Propodeum of Spathius marshi, sp. n. A, costula; B, lateral carina; C, dor-
sal carina; D, basal carina; E, fork of basal carina; F, dorsal area; G, areola;
H, area petiolaris.

THORAX.- The term thorax here means the morphological thorax plus
propodeum. Thoracic shape is of some value in species diagnoses. Occasion-
ally a specimen is found which at first appears to be radically different because
the thorax is strongly dorsoventrally compressed. These are presumably due
to unusually shallow pupation spaces and one must take care not to be misled
by them. However, S. calligaster, sp. n. is normally strongly flattened
(Fig. 17). Ihave found greatly flattened individuals of three additional species;
these are difficult to key because they are so distorted, but otherwise possess
the typical species characters. Occasionally, the form and development of the
prothoracic ‘‘lateral area’’ (see Nixon, p. 179) is of specific value. However,
the position of the ‘‘pronotal keel,’’ which Nixon found useful in the Old World
Spathius, is of little value in the Nearctic forms, being always at least partially
fused medially with the posterior margin of the pronotum.

Taxonomically important characters of the thorax include:

1. Mesonotal declivity as viewed from the side.

2. The degree to which the notauli are depressed in a furrow posteriorly.

3. Sternaulus* sculpture and form (for example, compare Figs. 5, 14 and 22).
The mesopleural disc immediately above shows considerable sculptural
variation in the Nearctic species, from smooth to strigose to granular to
rugose.

4. The scat carina crosses the mesosternum anteriorly behind the fore
coxae and continues dorsad onto the mesopleuron, forming the posterior
border of a roughly pentagonal prepectal area above the fore coxae at the
anterior end of the sternaulus (Figs. 5, 14, 17 and 22). The prepectal
carina is ‘‘incomplete’’ (Fig. 17) if its expression is obscured at any
point along the border of the prepectal area. Sculpture of the prepectal
area is also of some specific importance.

*
Equals mesopleural furrow in Nixon’s and Marsh’s terminology.

8 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

5. The number and form of the transverse carinae of the scutellar furrow is
of specific value in some cases. The scutellar disc is usually granular,
but may be rugose or smooth in certain species; the degree to which it
is flattened or conically raised (see Fig. 14) is also important.

6. Major carinae of the propodeum, and areas enclosed by them, are indicated
in Figure 4. In the European Spathius exarator, relative lengths and de-
gree of development of propodeal carinae have been shown to be highly
variable (Beyr, 1961); the same appears true for Nearctic species. How-
ever, the shape and sculpture of the enclosed areas, and within broad
limits the angle of carinae intersection, appear somewhat more reliable.
Also, the posteriolateral angles are occasionally produced to form dis-
tinct tubercles or projections which are very useful taxonomically.

LEGS.- The fore tibia bears a row of 10 to 50 stout spines on its anterior
edge. Their number and arrangement, while often difficult to see, is of specif-
ic value within a broad range, as is a row of as many as 9 spines at the apex of
the fore tibia. The outer apical lobe of the hind tibia (see Figs. 18 and 23)
usually also bears a row of spines whose number is of specific importance.
Frequently, relative lengths of the hind tarsal segments are also useful; how-
ever, leg shape is rarely of value except in the males of the trifasciatus group
which have the femora considerably swollen (Fig. 15). Presence or absence of
a dentiform projection ventrally at the base of the hind coxae may also be of
some specific value, although it is variously developed in nearly all the Nearctic
forms. The tibiae are banded in the majority of Nearctic Spathius; in some
species, however, the bands may be poorly demarcated, the tibiae appearing
unicolorous.

WINGS.- Terminology used in reference to Spathius fore wing venation is
shown in Figure 3. Presence or absence of infumated wing bands is important,
as are relative lengths of first intercubitus and second abscissa of the radius.
In addition, the subdiscoideal position, interstitial or not, and differences in
cell shapes and proportions, while subtle, are often useful. Males of one
Nearctic species, S. stigmatus, sp.n., have a well developed stigma present
in the hind wing (Fig. 13); it is of interest that such stigmal development also
occurs in several Old World species. Otherwise, hind wings of Nearctic spe-
cies are quite uniform, but variation occurs in the position of the postnervellus,
which if displaced basally divides the mediella into three abscissae; however,
this character is highly intraspecifically variable and not reliable.

ABDOMEN.- Numbering of the abdominal segments begins with the first
apparent segment (petiole) rather than with the first true morphological seg-
ment (propodeum) which for descriptive purposes is here included with the
thorax. As used here, the abdomen consists of two main parts, petiole and
gaster (segments 2-7). In lateral view, the petiole may be arched (Fig. 5) or
nearly straight (Fig. 22); length, while often showing considerable variation,
is sometimes of specific importance. Measurement of the petiole is made in
dorsal view and expressed as a proportion of the maximum dorsal length of
the following segment [tergite (2-3)]. Also, length of the petiolar sternite in
relation to total petiole length is of specific value in certain instances.

The first two gastral tergites are not separated dorsally, and I follow
Nixon and Marsh in referring to this segment as tergite (2-3). Presence or
absence of a thickened lateral margin just below the spiracles on tergite (2-3)
is a useful specific character for females; males uniformly possess this
thickened lateral margin, which may often be present on subsequent tergites.

Although petiolar sculpture is of little value (see Beyr, 1961), consisting
mostly of irregular longitudinal striation, sculpture of the gaster is an impor-

Matthews: Nearctic Spathius 9

tant character. Its form and expression may, however, vary somewhat both
within and between the sexes of a given species. Usually sculpture, when
present, is confined to the basal portion of tergite (2-3) but in S. calligaster,
sp. n. much of the gaster is coarsely granular. If sculpture occurs on ter-
gites 4 to 7, it is usually in the form of fine punctulation. The epipleura of
the tergites, defined as that part of the tergites ventral to the spiracles and
covering the lateral face of the abdomen, may also show fine punctate or
shagreened sculpture in some species.

Male genitalia of all known species were examined in the course of this
study, but on the whole are rather uniform and few useful characters were
noted. Differences in number, arrangement and relative lengths of the para-
mere setae, size and shape of basal ring, and relative length of aedeagal apo-
demes may prove to be of use in distinguishing males of some species.

SIZE, COLOR AND VESTITURE.- Because it is not influenced by head
position and abdominal distension, fore wing length (FWL), measured from
the free edge of the tegula to the apex of the wing, is given in the descriptions
aS a more reliable indication of size than body length (BL); for the same rea-
son, ovipositor sheath length (OSL)* is expressed as a proportion of wing
rather than of body or abdomen length.

For the most part, the chief diagnostic value of color lies in the arrange-
ments and patterns produced, such as the banded hind tibiae, so that attempts
to use exact color terminology would only make the descriptions more cumber-
some while adding little to their usefulness. Considerable intraspecif-
ic variation in body color may occur in many species and is specifically noted
in those cases.

Although Nixon made extensive use of vestiture characters, especially
relative length of dorsal hairs of the hind tibia, little variation of this nature
was found to occur in the Nearctic forms.

GENERAL REMARKS.- Many species of Spathius described by Ashmead
(1893) were based on specimens reared by A. D. Hopkins. The previous year
Hopkins (1892) had published a list of his bred parasites in Insect Life, using
Ashmead’s manuscript names. Hopkins’ original manuscript for this paper is
on file in the U. S. National Museum; it is identical to the published version
except that Hopkins’ specimen label numbers are listed next to the species
names. Association of specimens and numbers has been of crucial importance
in this revision since most of Hopkins’ specimens (and consequently Ashmead’s
types) have no labels other than this number, and in some cases several speci-
mens have the same number. Hopkins’ manuscript has thus often been the sole
means for determining which specimens Ashmead had before him and for as-
sociating specimens with host and rearing data. Associated insects, host trees
and other data (locality, date of emergence, etc.) are recorded on cards filed
by number in the U. S. National Museum. I have made extensive use of this
Hopkins card file as well, and have listed the Hopkins numbers in the species
discussions where they were sources of host or other information.

Much of the confusion so long surrounding the identity of the described
Nearctic Spathius species can be traced to Ashmead’s (1893) descriptions and
key. The major characters he used were body color, wing infumation pattern
and ovipositor length, the first two of which have proved totally inadequate.
Furthermore, his often hurried work resulted in considerable carelessness
and inconsistency. For example, his key couplet to S. unifasciatus

*
Ovipositor sheath length was measured rather than ovipositor length per se
because the apparent ovipositor length depends greatly upon angle of extrusion.

10 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

(= trifasciatus Riley) states ‘‘ovipositor twice the length of abdomen’’, a fact
contradicted by both his own formal description of the species and by examina-
tion of the type. At another point he states that S. laflammei Provancher has
an ovipositor twice its body length; even if Ashmead had not seen the type,
Provancher’s description clearly states ‘‘Tarriere presque aussi longue que
le corps’’ (Provancher, 1880, p. 164). .

Genus SPATHIUS Nees

Spathius Nees, 1818, Nova Acta Acad. Caesar. Leop. Carol. 9: 301.
Type: Ichneumon exarator Linnaeus. Monobasic.
Stenophasmus Smith, 1859, J. Linn. Soc. Lond. 3: 169.
Type: Stenophasmus ruficeps Smith. Monobasic.
Euspathius Foerster, 1862, Verh. naturh. Ver. preuss. Rheinl. 19: 236.
Emendation of Spathius.
Rhacospathius Cameron, 1905, Spolia zeylan. 3: 86. Type: Rhacospathius
striolatus Cameron. Monobasic.

Head subcubical, variously sculptured; occiput margined; ocellar triangle
with base usually longer than sides, rarely equilateral; first flagellar segment
at least as long as second, usually slightly longer; flagellum with 16 or more
segments; fore and mid tibiae with 10-55 spines along anterior edge and a row
of 4-11 at apex; outer apical lobe of hind tibia with a row of 0-9 spines; fore
wing with 3 cubital cells, the recurrent vein entering the second cubital cell;
subdiscoideus leaving first brachial cell above middle, occasionally inter-
stitial with discoideus; submediellan cell present in hind wing but short, second
abscissa of mediella always longer than first; hind wing of males occasionally
with stigma-like thickening at junction of costella, basella and radiella veins;
first abdominal tergite lengthened, petiolate, usually abruptly dilated at apex;
gaster smooth or variously sculptured, sculpture when present often confined
to tergite (2-3); tergite 7 with area of close longitudinal striation on basal
half, this area often hidden by overlapping margin of previous tergite; epi-
pleura of tergite (2-3) always fused in females, but separated in males*; ovi-
positor variable in length but always at least half as long as gaster; male
genitalia as in Figure 20.

Discussion. - The essential characters by which Spathius may be dis-
tinguished from all other Doryctinae are: (1) abdomen petiolate, the first
abdominal tergite lengthened and usually abruptly dilated at apex; (2) fore
wing with three cubital cells, the recurrent vein always received into the.
second cubital cell; and (3) the subdiscoideus always originating above middle
of first brachial cell.

The condition of the epipleura of tergite (2-3) - fused in females and
divided in males - is somewhat unique to Spathius. Separation of the epipleura
of tergite (2-3) in both sexes is characteristic of most other doryctine genera
(about 90%, Marsh, in litt.). The adaptive significance of fusion versus
division is not understood.

Additionally, the position and form of the stigma in the male hind wing of
some species is unusual, being a thickening at the junction of the costella,
radiella and basella veins (Fig. 13; see also S. erigone Nixon, 1943, p. 290).

*
Nixon (1943) apparently overlooked the separated epipleura of the males, stating
in his generic description that the epipleura are always fused.

Matthews: Nearctic Spathius 11

In most doryctines having stigmal development in the male hind wing (e.g.
Acrophasmus, Heterospilus s) the stigma is a large thickening along the costal
margin and the other veins are greatly reduced or absent (see figures in Marsh,
1965).

Until quite recently, Spathius and its relatives were recognized as a sepa-
rate subfamily, the Spathiinae. Both Nixon (1943) and Marsh (1965) have ade-
quately reviewed the history of the classification of Spathius and related genera,
particularly those of the stephaniscine complex. (For a review of the classifica-
tion and phylogeny of the entire Braconidae see Tobias, 1967.) In his compre-
hensive review of the Nearctic Doryctinae, Marsh (1965) has included Spathius
and all of the stephaniscine complex in his redefined Doryctinae, based upon
‘*similarities among these groups in wing venation, fore tibial spines, male
genitalia and general morphology’’. Tobias (1967) concurs, but regards the
group as a supertribe, Doryctina, in his subfamily Braconinae.

As the situation now stands, I am in general agreement with Marsh and
Tobias. However, having seen a considerable sample of unnamed Neotropical
doryctines, a large proportion having the typical ‘‘Spathius’’ facies, Iam con-
vinced that the present system still leaves much to be desired. It is apparent
that the spathiines have radiated extensively in the Neotropics, the diversity
being perhaps greater than that found by Nixon in the Old World Malayan
Archipelago, which he regarded as the center of distribution for the genus.

At present, there are some 7 species of Spathius described from the Neotrop-
ics (not including several from the West Indies which are discussed by Matthews
and Marsh, 1969). While much like typical Spathius, all differ in having the
recurrent vein received by the first cubital cell and the subdiscoideus arising
below the middle of the brachial cell; this venation is widespread among Neo-
tropical ‘‘Spathius’’. In addition, I have found at least three other distinct
venational types, including the typicalform. These anomalies can be dealt
with only after further study. For these reasons, conclusions on generic,
tribal and subfamily limits in this complex must await redefinition of the
Doryctinae based on more thorough knowledge of the world fauna, as classifi-
cations based primarily upon north temperate faunas are proving increasingly
inadequate. For the same reasons, I prefer to defer the question of the generic
affinities of Spathius. Both Nixon and Marsh regard Rhaconotus Ruthe as the
closest relative of Spathius. However, as Nixon pointed out, the blister-like
swelling on the front and middle femora and the unfused epipleurites of tergite
(2-3) characteristic of the Rhaconotus-Platyspathius complex are not at all
typical of Spathius.

Paleontology.- Four spathiine fossil species from the Tertiary have been
described - three from Rott by Statz (1936, 1938) anda fourth, Scudder’s
Ichneumon petrinus, noted by Brues (1910). Through the courtesy of Dr. Charles
L. Hogue, Senior Curator of Entomology at the Los Angeles County Museum, I
have been able to study Statz’s Spathius types. The wings of S. pedicularis are
hyaline and the specimen is beautifully preserved. The fossils of S. macroradi-
alis and S. longicornis are less well preserved.

~ Of these three, only S. pedicularis is a true spathiine, its venation agreeing
with that noted earlier for the majority of Neotropical species, viz., the re-
current vein entering the first cubital cell and the subdiscoideus ar arising below
the middle of the brachial cell. S. longicornis, although probably a doryctine,
exhibits the wrong petiole shape and venation for Spathius. S. macroradialis
appears to belong to the Opiinae, having the typical venation and petiole shape
of that group.

12 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

Scudder’s (1890, pl. 5, fig. 14) figure lacks sufficient detail for precise
placement, and I have not seen his specimen. Judging from his figure, the
petiole of Ichneumon petrinus does not appear to have the typical spathiine
shape. Brues is probably correct, however, in assigning it to the Braconidae.

In over 1,000 blocks of Tertiary and Cretaceous amber currently under
study, no spathiines have been found (L. Masner, in litt.).

Distribution. - Representatives of Spathius are described from every |
zoogeographic region. According to Nixon (1943) the greatest concentration
of species occurs in the Malay Archipelago, the group apparently originating
in that region. However, as noted above, there also appears to have been an
extensive spathiine radiation in the Neotropical Region. Nixon recognized only
6 species from the Palearctic Region, but was handicapped by not having access
to many of the types; more recently, Fischer (1966) has given a key to 12 spe-
cies occuring in Europe (one new) and at least two others have also beende-
scribed from the Palearctic Region since 1943. For the Nearctic Region I
here recognize 21 species. No Holarctic species are currently known. I have
examined the scanty Palearctic material available in the U. S. National
Museum, and although the possibility of holarctic distributions cannot be
totally ruled out at this time, none of the North American species will success-
fully run through Nixon’s or Fischer’ s keys.

Because of their parasitism of wood-boring beetles, the chance of accidental
introductions is ever present. Indeed, at least 4 foreign species (based on
Specimens in the U. S. National Museum and Canadian National Collection) have
been intercepted in infested wood items at American ports of entry. However,
none appear to have become established. One of these, Spathius vulnificus
Wilkinson, a parasite of bostrichid beetles from India, has been intercepted
at no fewer than 4 different ports. Another, Spathius exarator Linnaeus, a
parasite of anobiids from Europe, has been frequently recovered from imported
items.

EVOLUTIONARY TRENDS

Although in a few cases the Nearctic species could be construed to fit cer-
tain of Nixon’s (1943) groups of Old World Spathius (particularly the exarator,
elaboratus, fasciatus and labdacus groups), none possess the exact combination
of characteristics of these groups; where individual North American species
show resemblance to Old World species, this has usually been mentioned in
the species discussions. For purposes of showing relationships, however, I
have divided the Nearctic Spathius into the following species groups, which I
feel represent more or less natural groupings:

1. The brachyurus group: brachyurus Ashmead, brevipalpus, sp. n., marshi,
sp. n. and stigmatus, sp. n.

2. The calligaster group: calligaster, sp. n.

3. The longipetiolatus group: longipetiolatus Ashmead, elegans, Sp. n.

and evansi, sp. n.

4, The sequoiae group: sequoiae Ashmead, canadensis Ashmead and aphenges,
sp. n.

5. The pallidus group: pallidus Ashmead, parvulus, sp. n. impus, sp. n. and

comes, sp. nN.

The laflammei group: laflammei Provancher and benefactor, sp. n.

The trifasciatus group: trifasciatus Ashmead and brunneus Ashmead.

The simillimus group: simillimus Ashmead and floridanus Ashmead.

co -1 0

Matthews: Nearctic Spathius

releteletel Gttacking hosts in reteset etter entree reer etet ener eee eeeteeeteeeetseeteseeeresereseresererecerececeraterereteretetens
eeetetetes, CONIFEROUS trees SERN NNN NNN SCS Se sete latest tate,
ine ciel ot. as vetere i Ay attacking hosts in wotetee worn
neseenest tHE Sascesse DECIDUOUS trees SS eae
parasitic primarily genes wre cteteterets
upon SCOLYTIDAE ttttth = orocerecececee
- ° 4 BSesees = 0 6 66 6 6 4
nee e t is aESseenGs | 7} ee 2 een 207018
‘ ee fd aere, es pi os es ee va (Ve! ees
ae ot es sage sae Cac S meaaiiag anni Ss 32
gels ol eee es, oe ee ee oF s < ee ae:
orice ees = > {4 = i) = ea cS <x < o S Siege Pa a test si = a ze heaps rece
ss a ea a ee oe aS SSS eo ee 8 | ae © eee eae 8
_e@ @ as = e o j= uw Sy Same ee <x =a 4 — =< us ele.
6 = <x <x a Gas ee a c_-— any ay > eS <x — a eee.
Do a 128 7 a a jel <x Av2o es 4 co poe we ce ee tu ear S) n = co a
Seale way x Litbickdtte Ba os ae
: ee hee Po
3 4 if f ie
3 e 3 H ra 3
H 3 ‘ 3
ot Fa
3 f
f
- é
7 TRIFASCIATUS f fete 6
3 / GROUP GROUP io
PALLIDUS i LAFLAMME! ti, POE
8 sy
anne i carina below GROUP as and : Hes and
H em ax
i ee sine fiction cd ce
‘ swollen | sculptured og
an SE QUOIAE LONGIPETIOLATUS 2 oy,
uniformly 3 *3e
transversely } GROUP GROUP aie
strigose ee folie

C. weakly infumated

forewings E

\
%
\
‘

i
4
i
3
i

Figure 4A. Possible relationships within the Nearctic Spathius.

text.

vertex

broadened petiole

straight

ee eo G0 Sevecesoesoesoos Coe lecesoeoooNsEIECE Teg...

13

GROUP

eyes and palpi
reduced

Further explanation in

14 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

Figure 4A depicts my concept of the relationships among the North American
species of Spathius, based upon both morphological and biological characters.
Two main lines of descent are apparent, one being that taken by the brachyurus
group and the other including the remaining 17 species, within which several
distinct evolutionary trends can be recognized.

Members of the brachyurus group are united by the combination of the
reduced maxillary palpi, reduced and conspicuously hairy eyes, broad temples,
long malar spaces, first intercubitus much shorter than second abscissa of
radius, and reduced antennae, the flagellomeres approaching a subcubical shape.
The group is clearly specialized, as evidenced by the tendencies toward reduc-
tion in maxillary palpi, number of flagellomeres and eye size. The trend to-
ward reduction of the maxillary palpi reaches its extreme in S. brevipalpus,
which has apparently completely lost one segment in addition to having the
others highly reduced. The brachyurus group does not show clear affinities
to any Old World species nor to the remaining Nearctic species. Perhaps it
should be given subgeneric or generic rank; however, I prefer to defer such
designation until such time as the doryctine fauna is more completely known
and generic limits better defined.

Within the brachyurus group, S. brachyurus is apparently the least special-
ized, having the most nearly normal eyes and the greatest number of flagello-
meres; S. brevipalpus seems closely related to brachyurus, both possessing
the unique tubercles at the posterior-dorsal-lateral angles of the propodeum and
nearly identical body sculpture. Additionally, both are parasites of beetles in
pines. The remaining members, S. marshiand S&S. stigmatus, are both para-
sites of beetles in deciduous trees, but apparently represent separate evolu-
tionary trends. Of the two, stigmatus is probably the more specialized, having
the fewest number of flagellomeres (16-20) of any Nearctic species. Presence
of a stigma in the male hind wing, while unique among the Nearctic species, is
probably a convergent character, since similar stigma-like thickenings occur
sporadically in several Old World species; this character’s adaptive signifi-
cance is not understood. The unique combination of characters possessed by
S. marshi, particularly the form of the sternaulus (Fig. 5) and the rugose
scutellum, apparently have no parallel in any described Spathius; additionally,
the wings are totally hyaline, somewhat atypical for the genus.

The calligaster group is represented only by Spathius calligaster, an iso-
lated species of uncertain affinities; it is characterized by the strongly flattened
head and thorax (Fig. 17), coarsely granular sculpture on all but the apical
margins of tergites (2-3), 4 and sometimes 5, absence of spines on the outer
apical lobes of the hind tibiae, strongly thickened lateral margin of tergite
(2-3) and the typically interstitial subdiscoideus vein. The flattened body is
convergent with species of Nixon’s labdacus group but other characters do not
correlate. Biologically it is unique in attacking eucnemid beetles.

The longipetiolatus group includes the largest of the Nearctic species.
Except for isolated rearing records from various hardwoods and a single host
record (for elegans attacking an anobiid), nothing is known of the biology of this
group. The three included species are not closely related, and the group’s
affinities are uncertain although some characters are shared with the Palearc-
tic exarator group. Available evidence suggests that the longipetiolatus group
may be most generalized of the Nearctic groups and probably closest to the
ancestral spathiine stock, its members being characterized by relatively large
size, long ovipositors (the ? longipetiolatus is unknown), and elongate, straight
petioles (Fig. 22). The antennae of evansi have the most flagellomeres of any
Nearctic species (35-48); it is also one of the most widely distributed of the

Matthews: Nearctic Spathius 15

of the Nearctic species, occurring from the Rockies eastward.

Members of the laflammei, simillimus and trifasciatus groups appear to be
derived from the same basic stock, but have split into three distinct evolution-
ary lines. All three groups share the relatively wide vertex, a well-developed
row of 3-10 spines on the outer apical lobe of the hind tibia, and strongly banded
fore wings; additionally, they all attack beetles in deciduous trees. A dominant
evolutionary trend in the complex seems to be the acquisition of a relatively
wide vertex and broad temples; a general increase in body robustness is also
apparent.

The laflammei group seems to represent the basic stock of the complex,
with S. benefactor apparently the most generalized member, having the nar-
rowest vertex. It is sometimes hard to distinguish this species from members
of the pallidus and sequoiae groups. Both benefactor and laflammei attack
weevils and bark beetles in deciduous trees, benefactor being commonly reared
from elm.

The trifasciatus group has acquired at least two unique features: the pres-
ence of a strong longitudinal carina beneath the sternaulus (Fig. 14) and strongly
swollen femora in the males (Fig. 15). The two included species attack closely
related species of scolytids, trifasciatus being commonly reared from Scolytus
quadrispinosus in hickory and brunneus from §. muticus in hackberry.

As the third group of this complex, the simillimus group is characterized
by the very strong uniformly granular-strigose sculpture of the vertex and
temples and the usually rugose or wrinkled scutellar furrow carinae. Its two
members have become specialized upon a different host assemblage, attacking
either cerambycid or buprestid larvae almost exclusively.

All of the members of the sequoiae and pallidus groups have become adapted
to beetles attacking conifers, a major ecological shift, and within each group
certain secondary trends can also be recognized; the species comprising both
groups include some of the smallest and taxonomically most difficult members
of the Nearctic fauna. The sequoiae group is characterized by weakly infumated
indistinctly banded fore wings and a nearly smooth to delicately strigose vertex.
Of the three included species, S. aphenges is probably the most advanced, with
the shortest ovipositor of any Nearctic species (except parvulus), reduced head
sculpture, and specialization upon beetles attacking pine cones. ‘S. canadensis
and S. Sequoiae appear to be quite closely related and possibly subspecies, but
more data are needed to resolve that problem; both attack a similar host
complex, sequoiae being western in distribution and canadensis being primarily
eastern, the ranges overlapping in some parts of western North America.

Members of the pallidus group have the vertex uniformly transversely
strigose and have strongly banded fore wings. The three most closely related
members - impus, pallidus and parvulus - have relatively narrow vertex and
temples, and are parasitic upon scolytids in conifers. S. impus is apparently
restricted to Phloeosinus beetles attacking members of the Taxodiaceae and
Cupressaceae. S. parvulus and S. pallidus are closely related, parvulus being
a smaller species with somewhat reduced mesopleural and vertex sculpture,
and a shorter ovipositor; both attack scolytids and curculionids in various
species of pine. The remaining species, S. comes, is only tentatively included
here; it has broader temples, a longer ovipositor than the others, and has be-
come specialized on buprestids.

The origins and affinities of the Nearctic Spathius fauna are somewhat ob-
scure. The possibility exists that it could have been derived from some as
yet unknown Neotropical stock. However, based on material I have seen
there appears to be a distinct separation between the faunas of North America

9

16 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

and Mexico, and no species are known from the arid parts of the southwestern
United States. This gap may be an artifact of the efforts of collectors, but
probably represents a real situation, as the genus is primarily adapted to
forested regions. As noted earlier, Neotropical species have certain major
differences, especially in wing venation, so that Neotropical origin of the
North American species appears unlikely. Instead, the Nearctic Spathius
were probably derived from Palearctic elements and the possibility exists
that more than one invasion may have occurred.

BIOLOGY

Several isolated studies of the biology of various Spathius species around
the world have been made, but the information is too fragmentary for a full-
scale comparative consideration of this subject at the present time. Fora
summary of biological information on Old World species, reference may be
made to Nixon (1943). One paper overlooked by Nixon is that of Russo (1938)
which includes a discussion of the biology of Spathius rubidus Rossi. Since
1943, two excellent papers by Lyngnes (1955, 1960) on the European Spathius
exarator Linnaeus have appeared, which include descriptions of mating and
oviposition behavior, larval development, and functional morphology of the
female reproductive system; additional more recent papers include a study of
the Philippine S. fuscipennis Ashmead (Ishii, 1953) and a brief comparative
treatment of internal ovarian morphology of a Japanese species (Iwata, 1959).

For the Nearctic species, published biological information (other than host
records) exists only for S. benefactor, sp. n., reported under the name S.
canadensis Ashmead; under the description in this paper, work on the life his-
tory of this species (Kaston, 1937, 1939; Robert, 1951 a&b, 1960) is summarized
briefly. In the following paragraphs, mention will be made of only a few salient
general features regarding the biology of Nearctic Spathius.
wood-boring beetles, larvae of the genus Spathius are essentially separated in
definition from general predators only by consumption of a single host rather
than several (Doutt, 1959). In laboratory studies, species have been shown to
satisfactorily accept novel hosts, even of a different family than forms their
usual diet (Ayyar and Narrayanaswami, 1940). Furthermore, the numerous
hosts recorded for the more thoroughly studied species cause one to additionally
suspect that, in conformity with generally noted evolutionary trends among the
ectoparasitic Hymenoptera (see Townes, 1962), the main specializations of
Spathius are to the site at which the host feeds rather than to taxon of host.

Broadly, the Nearctic species may be divided into two groups: those which
attack beetles in coniferous trees and those which attack beetles in deciduous
trees (see Fig. 4A)*. Adaptation to elements within each of these broad ecologi-
cal zones apparently has been achieved on several independent occasions. For
example, species from four different stocks (stigmatus, laflammei, trifasciatus
and floridanus) are adapted to attacking beetles in hickory; in pine, three Spathius
stocks have converged upon the beetles (2 species each in the brachyurus, pallidus

Onl S. floridanus has been reared from both coniferous and deciduous trees, but
it exhibits a decided preference for the latter, attacking several different nut trees
(hickory, walnut, chestnut and oaks), and the single record of this species from bald
cypress requires further confirmation.

Matthews: Nearctic Spathius 17

and sequoiae groups). Within the broad division as to tree types, a second
grouping may be made, separating those Spathius species which are predomin-
antly parasites of Scolytidae and those attacking non-scolytid borers of several
families (see Fig. 4A); however, this division is much less clear-cut. This is
not unexpected, since the ovipositing female of each species presumably searches
for hosts in a particular ecological niche, there attacking a variety which happen
to meet her particular requirements. There are several cases of species attack-
ing both scolytids and curculionids (e.g. brachyurus, sequoiae, benefactor); on
the other hand, species attacking the larger borers of the families Cerambycidae
and Buprestidae (e.g. simillimus, floridanus) seem somewhat more restricted to
these groups. In general, species with shorter ovipositors tend to attack cambial-
and bark-inhabiting scolytids and curculionids, while those with longer oviposi-
tors are parasites of beetles which bore deeper in the tree, but there is not neces-
sarily a correlation between ovipositor length and host taxon. (Spathius aphenges
is unusual among the Nearctic species in apparently preferentially parasitizing
beetles attacking the cones of pine rather than tree trunk and branches.) Another
habit which appears to have independently evolved more than once in Spathius is
that of multiple oviposition in a given host. I have noted evidence for this situa-
tion in at least two North American species (brachyurus, floridanus; see ‘‘Biolo-
gy’’ for those species), and it is also recorded for S. vulnificus Wilkinson (see
Nixon, 1943). It will probably turn up quite commonly as the biologies of more
Spathius species become known.

Within the Nearctic Spathius, hosts have been recorded for all but three
species (longipetiolatus, evansi and marshi); these are summarized in the
accompanying host list. However it must be emphasized that our knowledge of
host relations in this genus is still quite incomplete.

Known beetle hosts of Nearctic Spathius species. Sources for the records
are given in the species descriptions. Asterisks indicate records of associa-
tion only; definite parasitization of the named host not established.

Anobiidae
Eucrada humeralis Melsheimer, . 23900. 2 eo laflammei Provancher
WI OUTS age iiel ke Ale fe a ee wo pi ar ee elegans, Sp. n.
Buprestidae
PDI Seas GORY ee: ehh vik «4. Bi Hew d Bee Ewa simillimus Ashmead
PCO ey seks ia Win he Ad puldcce Ke b Mildheknrd bias asin brunneus Ashmead
Wom Reais (WeDET) sage pk Se eee 6 ee Simillimus Ashmead
Charysopothris femoratat(Olivier). . i «a < «628 «ow wo floridanus Ashmead
Ceara CACC ION 6 ee bia owe Gk wee ee eee comes, Sp. n.
Re eS Sa ee ik at ey ee a hice EL Re ee ee floridanus Ashmead
Weianaridia filyoettlata (earvis):. 6 soo Mv ek oc ek ee ee comes, sp. n.
Cerambycidae
Catliadiurme aereum: Newman 600 ay oe CO ae ee simillimus Ashmead
epios ys Ma0ula Bayi s Vee See ee -. *calligaster, sp. n.
Leptura muta DiS Newiran i ioe to oe ee ee eke bee *calligaster, sp. n.
Neichnea Jaticornis Says ns Se a ae es . *stigmatus, sp. n.

Xylotrechus colonus (Fabricius)... ... ...08. .... . floridanus Ashmead

18 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970 -
Curculionidae
Macdalis.armicollis Say. i hk Share Se we ee benefactor, Sp. Nn.
Wi arise Gays 6456 Fe ES ae ek benefactor, sp. n.
Mi. iconsnicre Herd 2.46) 656 tee ee Oa benefactor, sp. n.
PE AOL Paste Wee a NS BR ee We BOR. stigmatus, sp. n.
laflammei Provancher
floridanus Ashmead
Pachylobius picivorous (Germar) .......... .. *brevipalpus, sp. n.
Pissodes approximatus Hopkins. ............ brachyurus Ashmead
pallidus Ashmead
Pe hemerensis Germat 2 A ae eS ‘ pallidus Ashmead
Py Sera Pern ee Se eS aed Se te 28 brachyurus Ashmead
parvulus, sp. n.
PS SOS S8e ase Sas CON Ne we Saas tie ew, Be sequoiae Ashmead
EFucnemidae
Melasis pectinicornis Melsheimer. ............ calligaster, sp. n.
Scolytidae
Carphoborus intermedius Wood ........2.-e.s-ee- *sequoiae Ashmead
Conophthorus coniperda (Schwarz)... ......sseee-s aphenges, sp. n.
ROR Ses goa eas oe ee i Nae era canadensis Ashmead
Cryptureus etomins Le Conte es ae es ee 8 Pa SRE Rs *canadensis Ashmead
Dendroctonus obesus (Mannerheim)..........ee. sequoiae Ashmead
DD. TPOmale Di Merimans oo ee ee es we ewe pallidus Ashmead
D, pseudorsigae HOprins. i. eee eee es te » - sequoiae Ashmead
Dryocoetes affaber (Mannerheim)............. *sequoiae Ashmead
Dp. nutoerapius (RalzZebure) ec Se brachyurus Ashmead
canadensis Ashmead
Hylureopinds ralipes (Bichholi) sk ee ee kes benefactor, sp. n.
Bop ay este ee es gia ca eee oe ee pallidus Ashmead
Eeepandicoe ie (EachBOry oa a ke te ete se pallidus Ashmead
DPA TS TO PS ea Nae ae Se A i ae as aphenges, sp. n.
Pods Emirs eo Se Re ek eo Be bY oak ke ee 8 canadensis Ashmead
Ips sp. Se a ee eee eee a ee a oe » .. -parvulus, sp. n.
Leperieinus.atuicatus (Say) se S e S benefactor, sp. n.
Orthotomicus caelatus (Eichhoff) .......... ; canadensis Ashmead
Pityophthorus balsameus Blackman........... canadensis Ashmead
Po MUPPAVanAG PlAaCKMIAn 2 We ee RO Se ee . . *Ssequoiae Ashmead
PURO EO PBS Sa ee ae ie ae a ee ate . . *aphenges, sp. n.
Phloeosinus canadensis Swaine .......+e-2e+e-eve0-ee-e impus, sp. n.
Bre Se a GR BO a es Re eee BRE 6 impus, sp. n.
Po) DURCCALNIS TSU a eee ce Bh SR ee me wm a Ss sequoiae Ashmead
ES ne ike ne. Loe, ee a ie a oe sequoiae Ashmead
SPO | ee ee a ee hgh eae ae Aue, ie asa gel emt impus, sp. n.
Phioeotribus puperinis LeConle . ok G6 ew shane ee kene aphenges, sp. n.
Polygraphus rufipennis (Kirby) ... . 2... «wis + we ws *sequoiae Ashmead

canadensis Ashmead

Matthews: Nearctic Spathius 19

Pseudohylesinus nebulosus (LeConte)........... sequoiae Ashmead
Scolytus abietse Dintinmn ors 5 eee A re ee aphenges, sp. n.
uo LeepeeGwa ine eign es aig eg ctioe Nile geyid get sequoiae Ashmead
5. multistriatus Maushans ) i ois 2 wi Saige benefactor, sp. n.
S. MUIR aids as etna oe Pe ee eh eg brunneus Ashmead
B. piceme4ewainey yeh ee ak Pa ee ee ie *sequoiae Ashmead

canadensis Ashmead
o> Qupdrisphiogte Bays will sal nella? wae Bey trifasciatus Riley
S. tugvioses Miles 402 2 ie eh et) RA ae benefactor, sp. n.
5. Venivdie etanten eri ae Cae ee Fee sequoiae Ashmead
PCOLV TRE GUY argh. Be ns RRO PUL 4 eee: ee sequoiae Ashmead
Teypodendronm: Mneatunt OUVIE? 3 a Poe Aw Rie *sequoiae Ashmead
Ay loterinus pollitus 4Say)s:.5 SSeS ih eee ee stigmatus, sp. n.

Diapause. For establishment of a properly functioning host-parasite rela-
tionship, the two species obviously must be not only geographically and ecologi-
cally, but seasonally coincident as well. Synchronization of development with
host is often accomplished in the parasitic Hymenoptera through the mechanism
of diapause, but in these groups it is particularly difficult to unravel the causal
factors initiating this state (see Doutt, 1959). Those Spathius studied appear to
attack only the ultimate or penultimate larval instars (Kaston, 1939), but more
than one generation per year may commonly be raised upon the same favorable
log, and the meagre information available indicates that different species may
be either univoltine or multivoltine (see Fiske, 1907; Kaston, 1937, 1939;
Robert, 1951a & b, 1960). Overwintering of most Nearctic species probably is
in the prepupal stage, as shown for S. benefactor in southern Quebec (Robert,
1960) and in Connecticut (Kaston, 1939).

Adult habits and longevity. Spathius is a genus of forested habitats, and
the free-living adults are practically never found in open or gravelly situations.
Adult females apparently spend most of their time crawling over bark in their
search for hosts, and probably have a very short preoviposition period following
emergence. There is every indication that the genus is normally diurnal, but
rarely females of brachyurus, canadensis and brevipalpus have been collected
at light.

Adults of both sexes in at least some species feed upon carbohydrates, and
while rarely collected from flowers I have often taken them at sources of
homopteran honeydew. In the laboratory, adults feed readily upon a sucrose
Solution or well-soaked raisins. Adults of the Indian species S. critolaus Nixon
have been kept alive up to 162 days in the laboratory (see review in Nixon, 1943)
but probably the adult life is much shorter under natural conditions.

Sexual dimorphism. Among the Nearctic elements of the genus, sexual
dimorphism is not particularly pronounced. It is perhaps most evident in
males of the trifasciatus group, which have strongly swollen femora; although
mating has not been observed, this adaptation might be involved in clasping the
female. Asa general rule, males are smaller and have the sculpture less well
developed, except on tergite (2-3) where the sculpture may often be more exten-
Sive than inthe female. Because extensive courtship displays have never been
observed in the parasitic Hymenoptera, it seems more likely that the banded
wings of both sexes function primarily in camouflage or mimicry. Held folded
over the body, they often give Spathius an ant-like appearance from above.

Courtship and mating. As in most other parasites of wood-boring beetles,
Spathius presumably mate shortly after the female’s emergence, probably
while still upon the host tree. This is no doubt an important isolating mechanism,

20 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

for it would decrease the chance of sympatric allotrophic species encountering
one another during this critical time in their life cycle. This feature of its
biology, coupled with the apparently rather stereotyped habits of searching for
hosts, has important implications for speciation in the genus and may help ex-
plain the existence of so many widely sympatric species in eastern North
America.

Resource partitioning. Of special evolutionary interest is the occurrence of
several Spathius species attacking beetles (often of the same species) in the
same host tree. For example, at least five species have been reared from
beetles infesting hickory, at least three of which have been reported for the
same host, Magdalis olyra Herbst. Although perhaps not the case in this
particular complex, in an analagous situation involving the ichneumonid genus
Megarhyssa, the three species were found to be almost completely separated
by selection of hosts at depths corresponding to the different ovipositor lengths
(Heatwole and Davis, 1965). And while recently a number of investigators have
doubted the validity of universally applying the competitive exclusion principle,
and there is little information on the biology of these sympatric broadly auto-
trophic Spathius, it is possible (but not necessarily probable) that mechanisms
exist by which the hosts resources are partitioned spacially and/or temporally.
For example, at least one species, S. stigmatus, is temporarily isolated,
attacking a different beetle assemblage in trees in more advanced conditions
of decay (see ‘‘Hosts’’ under stigmatus). Other mechanisms - such as different
seasons of adult activity, preferences for different stages of the host beetle or
different parts of the host tree, or simply different daily activity cycles -
probably also exist behind the apparent coexistence of several species.

Key to the North American Species of Spathius

FEMALES
1. Malar space long, 0.67-1.00x EH; eyes relatively small (Figs. 5, 7 and
Be a ee a ee es ee Gv 2
Malar space 0.33-0.67x EH, usually about half EH; eyes normal (Figs.
Pa Oia a Se ee a 5)

2. Ovipositor very long, at least twice as long as fore wing; gaster with faint
aciculation at base of tergite (2-3).
a 4, stigmatus, new species (p. 32)
Ovipositor not longer than fore wing; gaster smooth and shining, with no
{race of sculpture at base of tereite (2-3)... oe ee. 3

3. Scutellum rugose; vertex and temples smooth; ovipositor 0. 69-0. 96x FWL;
body in profile as in Figure 5...... 3. marshi, new species (p. 30)
Scutellum centrally smooth and flat; vertex and temples coarsely granular

tO. Yusulose: Oviposiior 0.37-U0.GOx PWin wk kt tee we oe

4. Maxillary palpi greatly reduced, at most slightly longer than EH; temples

broad, from 1. 0-1. 3x EH (Fig. 7).
SO a es we 2. brevipalpus, new species (p. 29)
Maxillary palpi normal, at least twice as long as EH; temples less broad,
from 0. 76-1. 04x EH (Fig. 8). 1. brachyurus Ashmead (p. 26)

Matthews: Nearctic Spathius 21

Do. sternaulus bounded below by a well developed broadly curved carina extend-
ing from prepectal carina to mesocoxal cavity (Fig. 14); scutellum coni-
eal in. protien wie ea eas Se ee. 6
Sternaulus not as above, typically a simple longitudinal furrow extending
from prepectal carina posteriorly for about 2/3 length of mesopleuron
and crossed by few to many variously developed ridges (Figs. 17 and 22);
scutellum at most slightly conical or projecting apically. ....... 7

6. Ovipositor 0.45-0.68x FWL; head and thorax testaceous; first intercubitus
and second abscissa of radius subequal in length; profile of thorax as in
Mioure 14.03 oe es a a ee 19. brunneus Ashmead (p. 70)

Ovipositor 0.87-1.20x FWL; head and thorax ferrugineous to piceous; first
intercubitus distinctly longer than second abscissa of radius (Fig. 29).
RE es ae ee 18. trifasciatus Riley (p. 66)

7. Head and thorax strongly dorsoventrally compressed (Fig. 17), the pronotum
with a distinct dorsal transverse swelling; tergite (2-3) entirely coarsely
granular (except extreme apical margin) and lateral margin thickened over
nearly entire length; subdiscoideus vein typically interstitial (Fig. 18).
oe Se ee 5. calligaster, new species (p. 35)

Head and thorax not strongly flattened, or if apparently so then other
characters not as above; lateral margin of tergite (2-3) at most thickened
on basal half; subdiscoideus arising from upper half of brachial cell.

8. Petiole straight, slender and elongate (Fig. 22), usually much greater than
1.5x as long as middle femur; ovipositor long, 1.05-2.1x FWL.... 9*
Petiole usually distinctly arched at base (similar to that shown in Fig. 5)
and more stout, always shorter than 1.5x middle femur. ....... 10

9. Ovipositor very long, 1.6-2.1x FWL; hind tarsomere 3 distinctly longer
than 5; hind basitarsus not more than 2.1x as long as second tarsomere;
antennae with 35-48 flagellomeres. . . 8. evansi, new species (p. 42)

Ovipositor 1.05-1.55x FWL; hind tarsomere 3 subequal or slightly shorter
than 5; hind basitarsus 2.1-2.9x as long as second tarsomere; antennae
with 30-38 flagellomeres; propodeum often much lighter than remainder
of thorax, distinctly contrasting even to the naked eye.

Ree ee OR ee eee 7. elegans, new species (p. 39)

10. Vertex entirely strongly granular-strigose and very deep (OED/OOD 0. 56-
0.78, Fig. 24), the sculpture extending undiminished to eyes and over
ocellar triangle, frons and entire temples nearly to malar space; scutellar
furrow Carinae Olen Iwreeular Or TusOse. a es ee i 11

Vertex not as above, usually narrower (OED/OOD 0.58-1.11) and either
entirely or partly smooth (at least adjacent to eyes) or, if apparently
completely sculptured, then this in the form of transverse strigosity with
little evidence of the granular element; scutellar furrow carinae usually
Simple And More Gf tes Stra ee ae ye ees 12

“The unknown female of longipetiolatus Ashmead from Florida will presumably also
run out to this point in the key.

22 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

11. Gastral tergites 4-6 each with a well developed broad transverse punctu-
late band anterior to the row of setae; mesopleural disc above and pos-
terior to sternaulus evenly granular to scaly-reticulate; outer apical
margin of hind tibia with a row of 4-6 spines, typically 5.
boop brags Wi a CA aa ee, Sis 20. simillimus Ashmead (p. 72)

Gastral tergites nearly smooth and shining, with at most weak scaly-
reticulate sculpture on about basal half of tergite (2-3); mesopleural
disc above sternaulus with variously expressed strigosity; outer apical
margin of hind tibia with a row of 2-4 spines, typically 3. |
feat uti Gy SG Chg RD arte Rae See Ge EOE 21. floridanus Ashmead (p. 75)

12. Ovipositor very short, 0.32-0.46x FWL; vertex smooth and shining; ratio
of petiole length to dorsal length of tergite (2-3) 1.22-1.30; fore wing
subhyaline (Fig. 26); western North America.

USA. Wg Oise ie ERE wc ee eer tw 11. aphenges, new species (p. 51)
Ovipositor usually greater than .45x FWL; if shorter, then vertex at least

partly sculptured; ratio of petiole length to dorsal length of tergite

(2-3) variable, 1.26-2.00; fore wings variously infumated...... 13

13. Ovipositor 0.91-1.57x FWL; vertex strongly transversely strigose of uni-
form intensity from ocellar triangle to occipital carina, becoming granu-
lar at eyes; TW/EH 0. 63-0. 78, and OED/OOD 0. 64-0. 82.
ebook ate st eieasatss Serta a 15. comes, new species (p. 59)

Ovipositor shorter, 0.33-1.09x FWL; if ovipositor is greater than 0. 91x

FWL then either vertex is only faintly strigose, becoming smooth at
eyes, or temples and vertex are narrow, TW/EH less than 0. 63 and
OED/OOD greater than 0.85

14. Outer apical lobe of hind tibia with a row of 6-8 conspicuous spines (Fig. |
23); mesopleural disc above sternaulus smooth and shining; body length,
exclusive of ovipositor and antennae, 3.5-6.1 mm; antennae with 29-42
Lia emOmeres sia Ga eae ON 17. laflammei Provancher (p. 64)

Outer apical lobe of hind tibia with at most 6 spines, usually 2-5; meso-
pleural disc variable; smaller species, body length 1. 7-4.3 mm; anten-
Naeowitn 20-32 fiacellomieres 25 oi. ke ieee babi ee 15

15. Fore wings weakly infumated to subhyaline (Fig. 28), rarely with complete
dark transverse bands; vertex from nearly smooth and shining to deli-
cately transversely strigose; head and thorax usually dark, ferrugineous
to piceous; eyes gray, rarely weakly iridescent greenish in white light.

Fore wings with two dark well defined transverse bands separated by a
hyaline band through the basal 1/3 of stigma (Figs. 28-31); vertex vari-
able, sometimes nearly smooth, but often with well developed transverse
strigosity; head and thorax testaceous to ferrugineous; eyes typically
bright iridescent greenish in white light. .............00-. 17

16. Ovipositor 0.45-0.65x FWL; tergite (2-3) usually with a distinct area of
sculpture on about basal 0.37-0.45, and occasionally with additional
sculpture on middle third of tergite and/or weak transverse punctulate
bands on some subsequent tergites; body length 2.0-3.8 mm.

Ri ss inte cet Part eet Shia ete ROOM EL 10. canadensis Ashmead (p. 47)

Matthews: Nearctic Spathius 23

Ovipositor 0. 60-0. 80x FWL; tergite (2-3) either nearly smooth or with at
most weak mostly obscured sculpture on about basal 0. 38 of tergite;
body length 2.5-4.1 mm; western North America.

L praia 6 OR ARS aerae ay failee will deal 5 che Sty 9. sequoiae Ashmead (p. 44)

17. Vertex very narrow, OED/OOD 0.80-1.11; outer apical margin of hind
tibia with at most 4 rather inconspicuous spines, typically 2-3, situated

anteriorly (Fite 330 ee lee A ee Ee A ae 18
Vertex broader, OED/OOD 0. 70-0. 83; outer apical margin of hind tibia
With arow OF S-6 S0IBOS. 4 iwi te G Ree eS Ree... 19

18. Ovipositor 0.35-0. 64x FWL; tergite (2-3) with sculpture present on about
basal 0.65-0. 75; vertex striae delicate, tending to bend medially toward
ocellar triangle and blend into its granular sculpture (Fig. 25).
er me ae pray PV t ae pe: 13. parvulus, new species (p. 55)

Ovipositor 0.50-1.02x FWL; tergite (2-3) with sculpture on at most about
basal 0.40 of tergite, remaining tergites smooth; vertex striae more
or less transverse, not notably indented medially.

i de Seu Gratien te eras armaeds 2 14. impus, new species (p. 56)

19. Body color wholly tawny yellow to dirty brown, the gaster beyond basal
2/3 of tergite (2-3) occasionally slightly darker or with darker bands;
vertex more or less evenly transversely strigose; mesopleural disc
with delicately swirled strigosity above sternaulus.

Mee ape eee eee a Seay Ser a eee Gene 12. pallidus Ashmead (p. 53)
Body color shining ferrugineous, with about apical half of petiole to

middle of tergite (2-3) distinctly lighter, suffused with yellow or

rufo-testaceous; vertex anteriorly delicately transversely strigose and

shining; mesopleural disc above sternaulus smooth and shining or occas-

ionally weakly scaly-reticulate.

APR ROI E Tia RARE Tope eas ne are a 16. benefactor, new species (p. 61)

MALES

1. Scutellum flat, centrally smooth; propodeum with a prominent tubercle
projecting from each posterior-dorsal-lateral angle; vertex, temples
and mesopleura rugulose...... 1. brachyurus Ashmead* (p. 26)
Scutellum granular or rugose and at least weakly convex; posterior dorsal-
lateral angles of propodeum not produced; sculpture of vertex and temples
ER PO Tae eR, ine MAL aet Rke ah a we a, Wicd Beer FORE: “ghia: SRN pice. Whee OR lk 2

2. Hind wing with a well developed stigma-like thickening at the junction of
the costella, basella and radiella veins (Fig. 13); eyes small, the temples
broad, TW/EH 0. 90-1.04; antennae much shorter than body.
gh Ay ta shale gts Mt Nea ie agen Pa teal ae 4. stigmatus, new species (p. 32)
Hind wing not as above, if apparently with a weak stigma-like thickening
then TW/EH less than 0.90 and antennae about as long as body... . 3

3. Scutellum rugose; gaster smooth and shining; body in profile as in Figure 5,
the propodeum with a distinct tooth-like projection due to the strongly

*
The unknown male of brevipalpus, sp. n. will presumably also run to this point.

24

Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

protuberant carina separating the areola and area petiolaris.

ee aa oe a a 3. marshi, new species (p. 30)
Scutellum evenly scaly-reticulate to granular; gaster variable but often

with distinct sculpture at least at the base of tergite (2-3); prepedeum

NOU AS: ADOVE. 186 es ee. ee ee eee

Sternaulus bounded below by a well developed broadly curved carina extend-
ing from prepectal carina to mesocoxal cavity (Fig. 14); femora strongly
swollen (Fig. 15); scutellum conical in profile............

Sternaulus not as above, typically a simple longitudinal furrow extending
from prepectal carina posteriorly for about 2/3 length of mesopleuron
and crossed by few to many variously developed ridges (see Figs. 17 and
22); femora not particularly swollen (Fig. 16); scutellum at most ‘Slightly
conical Or projecting apicallys , 7 ee A ee

First intercubitus distinctly longer than second abscissa of radius (Fig. 29);
area where notauli converge rugose and strongly depressed between the
lateral lobes of mesonotum, ....... 18. trifasciatus Riley (p. 66)

First intercubitus sub-equal to second abscissa of radius; area of notauli
convergence shallowly depressed and not strongly rugose.

Py ee be rts Ge ee 19. brunneus Ashmead (p. 70)

Head and thorax strongly dorsoventrally compressed (Fig. 17); at least
tergites (2-3) and 4 uniformly granular (except apical margin).
a ee 5. calligaster, new species (p. 35)

Head and thorax not strongly dorsoventrally flattened; well-developed
granular sculpture rarely covering more than basal 3/4 of tergite (2-3).

i emer Fiala mao Jetn ee otuee( Foam Miro! eaves Sa Sal Seek ekaods( Saini Herat! Sabai Wkone’ Ry lot Teebedt boltinet ivciet Sobiea Meck? Sein! Seman Cama Setricsa Sota vie Salen! Vikaakhy lity Timea, jamais mY Maes |

Petiole elongate, slender and nearly straight (Fig. 22), longer than 1. 6x

lenetheot Middle TeMur s Oe ee ai Ga 8
Petiole shorter, usually distinctly arched at base (similar to that shown in
Fig. 5), always less than 1.55x length of middlefemur. ....... 10

Antennae with 22-25 flagellomeres; tergite (2-3) with at most weak acicula-
tion at extreme base; mesopleural disc above sternaulus smooth.
ee ee 6. longipetiolatus Ashmead (p. 38)
Antennae with more than 25 flagellomeres; tergite (2-3) with well-developed
sculpture on at least basal half; mesopleural disc above sternaulus granu-
fA WO PUSOSOE a ae eee ae ee ko eee eae gos 9

Vertex (and often frons) smooth, shining; sculpture on middle area of
tergite (2-3) often partially obscured, but when present, always much
less strongly developed than that of basal half of tergite and separated
from it by a narrow smooth space; mesopleural disc above sternaulus
typically scaly-reticulate to irregularly wrinkled.

Pe ea a i 7. elegans, new species (p. 39)

Vertex usually with faint transverse strigosity anteriorly; sculpture of ter-
gite (2-3) covering about 3/4 of tergite and uninterrupted at middle by a
smooth space, gradually diminishing in intensity apically; mesopleural
disc above sternaulus uniformly granular.

Oe ee a a 8. evansi, new species (p. 42)

10.

a.

12,

Os

14,

15.

16.

a

18.

Matthews: Nearctic Spathius 25

Vertex entirely prominently granular-strigose and very wide (OED/OOD
0.56-0.78, Fig. 24), the sculpture extending undiminished to eyes and
over ocellar triangle, frons, and entire temples nearly to malar space;
scutellar furrow carinae often irregularly wrinkled. ........ 11

Vertex not as above, usually narrower (OED/OOD 0.60-1.11) and either
all or partly smooth (at least adjacent to eyes), or if apparently complete-
ly sculptured then this in the form of transverse striation with little evi-
dence of the granular element; scutellar furrow carinae usually simple
andmore Or Lese Stale ee Pa eh ae eee. cae 12

Gastral tergites 4-6 each with a well developed broad transverse punctulate
band anterior to the row of setae. . . 20. simillimus Ashmead (p. 72)

Gastral tergites nearly smooth and shining, with at most scaly-reticulate
sculpture on about basal half of tergite (2-3).

oO UN Rr setae ents 21. floridanus Ashmead (p. 75)
Species occurring west of 100th meridian, 0°) oo ee Ve a 13
Species occurring east of 100th meridian or a eae 16

TW/EH 0.50-0.67; eyes iridescent greenish in white light; fore wings with
complete lightly infumated transverse bands (Fig. 30); vertex sculpture,
when present, as in Figure 25. ...13. parvulus, new species (p. 55)

TW/EH 0. 67-0. 85; eyes gray, at most faintly iridescent; fore wings hyaline
to weakly infumated in area of radial cell (Figs. 26 and 27). ..... 14

Frons nearly smooth, with faint traces of incomplete transverse striation;
space posterior to sternaulus smooth to faintly scaly-reticulate.
NS ah, CEng en oy Setar nn 11. aphenges, new species (p. 51)
Frons wholly delicately transversely striate, this sculpture often continuing
dorsally onto vertex; space posterior to sternaulus with about 2-6 closely
spaced gently curved loncitudinal ridees yu yer) ee 15

Ratio of lengths of hind tarsus to hind tibia 0. 66-0. 88; mesopleural disc
above sternaulus delicately strigose. .. 9. sequoiae Ashmead (p. 44)

Ratio of lengths of hind tarsus to hind tibia 0.84-0.96; mesopleural disc
above sternaulus scaly-reticulate. . . 10. canadensis Ashmead (p. 47)

Vertex distinctly delicately to strongly transversely strigose of more or
less uniform intensity from ocellar triangle to occipital carina. ... 17

Vertex nearly smooth, or if delicately strigose this only on about anterior
half, the part adjacent to occipital carina becoming smooth. ..... 20

Sculpture of tergite (2-3) covering no more than basal half of tergite (rarely
faint scaly-reticulation beyond middle of tergite); remaining tergites
SIO Oa ee SEO Lyre in Bae agtny Wrage mating leat abi Ms RD ane a ara en ges ce 18
Tergite (2-3) with distinct sculpture developed beyond middle, this usually
less extensive and of less intensity than sculpture on basal portion; re-
maining tergites (especially tergite 4) often with distinct punctulation or
scaly-reticulation anterior to the setae, ey. ee ee ee ok 19

Temples wide, TW/EH 0. 63-0. 88; hind tarsomere 2 distinctly longer than
tarsomere 5 (not including claw).
a eM eb a Sk sg ai i ee 15. comes, new species (p. 59)

26 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

Temples narrow, TW/EH 0.50-0. 63; hind tarsomere 2 subequal to tarso-
mere 5 (not including claw). ...... . 14. impus, new species (p. 56)

19. Vertex narrow, OED/OOD 0. 80-1. 00 (Fig. 25).
by pera eae vat Oe es aC Aes AS 13. parvulus, new species (p. 55)
Vertex slightly broader, OED/OOD 0. 70-0. 81.
Bei Sd nie REO AG hs eye eks 12. pallidus Ashmead (p. 53)

20. Outer apical margin of hind tibia with 5-8 strong spines, usually 6 or 7
(Fig. 23); mesopleural disc smooth immediately above sternaulus; ver-
tex wide, OED/OOD 0.58-0.78. . . .17. laflammei Provancher (p. 64)
Outer apical margin of hind tibia with 2-5 spines; mesopleural disc sculp-
fUPre SNe Verter Wid VaTIa Died ses ele PERRO! Be 21

21. Eyes usually bright iridescent greenish in white light; mesopleural disc
above sternaulus smooth (rarely weakly scaly-reticulate); fore wings
distinctly weakly banded, rarely subhyaline............4.-. 22

Eyes dark gray; mesopleural disc above sternaulus scaly-reticulate to
delicately strigose; wings typically hyaline, at most weakly infumated in
areaco radigl cals ong iws ee Baek 10. canadensis Ashmead (p. 47)

22. Vertex narrow, OED/OOD 0.85-1.11; head and thorax predominantly dirty

VEROW ea ae A ee ee ee aS 14, impus, new species (p. 56)
Vertex wider, OED/OOD 0. 67-0. 88; head and thorax predominantly dull
ferrugineous............ 16. benefactor, new species (p. 61)

1. Spathius brachyurus Ashmead
(Figs. 8, 10)

Rhopalophorus fasciatus Provancher, 1886, Naturaliste can. 16: 129. (nec
Spathius fasciatus Walker, 1874, Cist. Ent. 1; 307.) [Type locality:
Ontario: Ottawa (lectotype female in Laval Univ. , Quebec) ].

Eustalocerus fasciatus (Provancher), 1888, Naturaliste can. 18: 378.

Spathius brevicaudus Ashmead, 1892, in Hopkins, Insect Life 4: 258. Nomen
nudum,

Spathius brachyurus Ashmead, 1893, Can. Ent. 25: 71,73. [Type locality:
West Virginia: Morgantown (A. D. Hopkins) (lectotype female in USNM,
Type No. 2151)].

Spathius dignus Muesebeck and Walkley, 1951, Agriculture Monogr. 2: 169,
new name for fasciatus Provancher, New synonymy.

Females: Body length, 3.0-5.3 mm; FWL, 2.3-3.4 mm; ovipositor
sheath length, 1.0-1.9 mm. Other measurements and ratios in Table 1.

Head: Antennae short, about as long as FWL; flagellomeres 19-25, those
beyond the first subcubical, less than 2x as long as wide, each with a corona of
4 evenly spaced erect setae about 2x as long as all other flagellar setae; eyes
sparsely covered with short setae; face coarsely granular to rugose; frons
granular to rugose-reticulate, becoming granular in area of ocellar triangle;
vertex granular adjacent to ocellar triangle, becoming rugose striate posteri-
orly; temples rugose striate to coarsely granular; malar space and ocellar
triangle finely granular; mandibles striate below at base, the tips dark; maxil-
lary palpi slightly reduced, extending to about posterior border of fore coxae.

Matthews: Nearctic Spathius 27

Thorax: Lateral areas of prothorax distinct, well developed, delimited by
a distinct carina with several cross carinae; remainder of propleura coarsely
granular; notauli deeply impressed anteriorly, converging to a slightly depressed
rugose-reticulate area; mesonotum abruptly declivous, the lobes evenly granu-
lar, the middle lobe often with a weak medial longitudinal furrow; scutellar fur-
row with 4 to 7 cross carinae, often with some more strongly developed than
others; scutellum flat, centrally smooth, the posterior margin sometimes granu-
lar; mesopleural disc granular to rugose; sternaulus most strongly impressed
posteriorly, with 7 to 12 cross carinae superimposed on a granular background,
an irregular carina sometimes developed along the posterodorsal margin; pre-
pectal carina complete; prepectal area smooth to weakly granular; dorsal areas
and sides of propodeum coarsely granular to reticulate; all propodeal carinae
well developed and prominent, the basal carina about equal to its forks; areola
and area petiolaris crossed by several transverse carinae; posterior-dorsal-
lateral propodeal angles distinctly produced to form well developed tubercles;
fore tibia with 12-16 spines in an irregular single row and a row of 6 at apex;
outer apical lobe of hind tibia with a row of 3-4 rather elongate spines; hind
coxae angled basally below, ventrally coarsely granular with some oblique
striations dorsally; fore coxae smooth; middle coxae granular dorsally, smooth
ventrally; second abscissa of radius decidedly longer than first intercubitus; sub-
discoideus arising from upper half of brachial cell; tegulae smooth.

Abdomen: Petiole arched at base, longitudinally striate dorsally; gaster
entirely smooth and polished except ultimate tergite which bears longitudinal
striations on basal half (these usually not visible unless tergites are distended)
and a granular ventral apical margin; tergite (2-3) without thickened lateral
margins, but occasionally with faint aciculation at extreme base.

Color: Body color varying from uniformly testaceous (S. dignus) to dark
ferrugineous or castaneous; in some individuals, propodeum, petiole and most
of tergite (2-3) are distinctly lighter than remainder of body; legs usually con-
colorous with body, though in some specimens lighter, especially basal seg-
ments; tibiae not or indistinctly banded at base; clypeus, basal antennal segments
and mandible bases sometimes suffused with dirty yellow; ovipositor sheath tips
and apical 6-8 flagellomeres darker than respective basal portions, often brown;
flagellomeres each with a narrow darker apical band; fore wing as in Figure 10,
wings of males and lighter individuals approaching a subhyaline condition; costa,
media and basal veins and stigma brown, remaining veins light brown or tan.

Males: Agree with females in all essential features except for the laterally
thickened margin of tergite (2-3) and flagellum more slender, the first 5 flagel-
lomeres more than 2x as long as wide; intensity of body sculpture, especially
of vertex and temples, sometimes reduced; color variation also similar to that
of females.

Remarks: Through the courtesy of Dr. René Béique, I have examined
Provancher’s type of Rhopalophorus fasciatus* (renamed Spathius dignus by
Muesebeck and Walkley, 1951) and see no justification for retaining it as a
separate species. It agrees with the typical S. brachyurus in every respect
except that it is a very light-colored individual, being wholly testaceous except
for a darker gaster beyond tergite (2-3), but falls readily within the range of

“The lectotype of Rhopalophorus fasciatus bears a blue label with the handwritten
number 701 and a yellow label with the printed number 1281, and a lectotype label by
P.M. Marsh, Feb. , 1965. The abdomen is broken off and glued to the blue label. See
also Gahan and Rohwer, 1917, and Shenefelt, 1968.

28 Contrib. Amer. Ent. Inst. , vol. 4, no. 5, 1970

color variation exhibited by the material of S. brachyurus before me. Since
fasciatus is preoccupied by Walker (1874), Spathius brachyurus Ashmead be-
comes the oldest available name for the species, and S. dignus Muesebeck and —
Walkley falls as a synonym.

Following his description of S. brachyurus, Ashmead states ‘‘bred Nov. 10,
1890, and March 15, 1891, from Dryocoetes autographus living under bark of
dead Abies excelsa ...’’ (see also Hopkins, 1892). According to the original
Hopkins manuscript on file in the U. S. National Museum, specimens matching
these data bear the numbers 5139 and 167 respectively. Inthe U. S. National
Museum collection are two pins bearing the label ‘5139 Hopk. W. Va.’’* One
has four specimens of S. brachyurus, each on a separate point, and a cluster
of cocoons beneath on the pin. The other bears two specimens, each on a sepa-
rate point. The former has two additional labels, ‘‘Spathius brachyurus Ashm.
2’? in Ashmead’s writing and a red label ‘‘Type No. 2151 USNM’’. Of the four
original specimens, one is completely gone, a second broken and the other two
females in good condition. I have placed a red ink mark near the base of the
point of one of the good females, and here designate this female as the lectotype.
The other specimens having the number 5139 have been labeled as paralectotypes.

The number 167 is on the label of one specimen of Spathius in the U. S.
National Museum collection; however, this specimen is definitely not S. brachy-
urus, but S. canadensis. Also in the U. S. National Museum are one female,
one male, and two cocoons, all ona pin of S. brachyurus bearing the number
2442; this number is also on Hopkin’s original manuscript, with the same host
data except that the adults emerged April 6, 1891. Presumably Ashmead must
also have had these before him when he described 8, brachyurus, even though
he did not specifically mention any males in his description; thus, I am designat-
ing these also as paralectotypes.

In summary, 3 pins bearing a total of 7 intact specimens (699, 10) are be-
lieved to have been part of the original material on which Ashmead based his
description of S. brachyurus; one of these has been designated as lectotype and
the others as paralectotypes.

Diagnosis: This species may be distinguished from its closest relative,

S. brevipalpus, sp. n., by head shape, especially the less broad temples (Fig. 8),
by the more elongate maxillary palpi, and by having the gastral setae arranged

in a more or less distinct row on each tergite as opposed to the scattered
arrangement in two or three rows per tergite characteristic of brevipalpus.

Both may be readily separated from all other North American Spathius by the
combination of propodeal tubercle, flat scutellum, polished gaster and coarsely
sculptured head. :

Distribution: Eastern North America north of North Carolina and Tennessee.
I have seen specimens from Connecticut, Massachusetts, New Hampshire, New
York, North Carolina, Ohio, Rhode Island, Tennessee, Virginia, West Virginia,
Wisconsin, and Quebec, Canada. Also, in the U. S. National Museum is one
broken specimen labeled from the Chiricahua Mountains of Arizona; this is
possibly a labeling error.

ithe Hopkins manuscript has two numbers 5139. The first is indicated as bred
from a curculionid under Pinus inops bark, Sept. 28; the second bears the data cited in
Ashmead’s description; however, this second number was originally typed as 5189, with
the 8 subsequently written over as a 3. I suspect the first 5139 should then have been
changed to 5189.

Matthews: Nearctic Spathius 29

Biology: Except for host records, nothing is recorded for this species.
However, a cluster of 7 cocoons pinned with an equal number of adults (699,
1c’) from Branford, Connecticut (CAES), indicates that S. brachyurus, like
S. floridanus, has multiple oviposition in a single host. -

Hosts: Published host records for S. brachyurus include Pissodes strobi
(Peck) (Curculionidae) in pine and the scolytids Dryocoetes autographus s (Ratze-
burg) in spruce and Phloeosinus dentatus (Say) in red cedar. A summary of the
literature references to these hosts may be found in Bushing (1965). The last-
named host is probably erroneous, the original record (Riley and Howard, 1890)
referring to a manuscript species, Spathius abdominalis Riley,* and there are
no later records to substantiate it. No host records were published for S.
dignus.

The only new host record is Pissodes approximatus Hopkins in Scotch pine
from Virginia (Hopkins No. 10099t1). Host trees from which S. brachyurus
has been reared include Picea spp., Abies spp. and Pinus spp. (mostly stro-
bus); a series of females from Wood Co., Wisconsin (RDS), is labeled from —
light trap.

2. Spathius brevipalpus, new species
(Bigs, 7, 11)

Holotype. - female, North Carolina: Salisbury (J. S. Pinckney). [USNM
Type No. 70498].

Description of type female: Body length, 3.3 mm; FWL, 2.2 mm; oviposi-
tor sheath length, 1.2 mm. Other measurements and ratios in Table 1,

This species is very close to S. brachyurus and may be compared to it as
follows:

Head: Sculpture as in S. brachyurus; face in profile distinctly more pro-
tuberant (Fig. 7); antennae about as long as FWL, inserted on a distinct frontal
prominence; flagellomeres 19, those beyond the first subcubical, less than 2x
as long as wide and lacking a corona of distinctly longer setae; eyes smaller,
resulting in a longer TW and MS (see Table 1); maxillary palpi (Fig. 7) greatly
reduced, total length not longer than EH; hypostomal carina highly protuberant
to form a strong lip.

Thorax: Like S. brachyurus except as follows: lateral area of prothorax
less distinctly delimited posteriorly; scutellar disc delicately granular at mar-
gins; mesopleura and sternauli uniformly very coarsely granular, sternauli
lacking distinct cross carinae; fore tibia with an irregular row of about 15
spines and a row of 4 spines at apex; outer apical lobe of hind tibia with 3 spines;
basal carinae of propodeum about half as long as its forks, all propodeal cari-
nae strongly developed.

Abdomen: As inS. brachyurus except tergites have more setae, in two or
three irregular rows per tergite, as opposed to the single obvious row per ter-

vRiley’s manuscript species was never formally described and is therefore a nomen
nudum. Indications in the Hopkins Card File in the USNM suggest that this species was
conspecific with S. brachyurus Ashmead; this may have been the basis for its host
record being transferred to that species, first by Chamberlin (1939).

30 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

gite in S. brachyurus; base of tergite (2-3) weakly aciculated.

Color: Body uniformly testaceous; antennae uniformly testaceous, flagel-
lomeres lacking a darker apical band; wings hyaline (Fig. 11), no trace of a
band, all veins and stigma uniformly light brown or tan (second intercubitus
and intersection of recurrens, first cubitus and first intercubitus not effaced).

Variation: Paratype females. Major variation is in color, the two para-
types being more darkly pigmented than the type, the one from Raleigh, North
Carolina, almost uniformly brown. The spines on the fore tibiae are difficult
to see and count accurately.

Male: Unknown.

Paratypes: North Carolina: Raleigh, 19, 1 Sept. 1947, light (M. W. Wing)
USNM]. South Carolina: Spartansburg, 19, Hopk. No. 11209a3 (T. E. Snyder)

[
[ USNM].

Remarks: This unique species is readily distinguished from all other
North American Spathius by the greatly reduced maxillary palpi which are the
basis for the specific name, brevipalpus.

Distribution: Aside from the type material, two additional broken female
specimens are in the U. S. National Museum collection, one from Durham,
N. C., the other from Jacksonville, Texas. This suggests that S. brevipalpus
probably has a range covering the southeastern United States west to Texas.

Hosts: The paratype with Hopkins No. 11209a3 was bred from “red top
pine’’; a list of associated insects is under this number in the Hopkins card
file at the U. S. National Museum. The label on the specimen from Durham,
N. C., states ‘‘ex Pachylobius picivorous’’, a curculionid, a record which
should await further confirmation.

3. Spathius marshi, new species
(Figs. 4, 5, 9)

Holotype.- female, Maryland: Patuxtent [= Patuxent] Wildlife Ref. (P. M.
Marsh). [USNM Type No. 70499].

Description of type female: Body length, 3.3 mm; FWL, 2.6 mm; oviposi-
tor sheath length, 1.8 mm. Other measurements and ratios in Table 1.

Head: Antennae slender, about 1.3x FWL; flagellomeres 22 right side, 23
left side, each with apical corona of 4 erect setae, about 2x as long as other
flagellomere setae; face smooth with transversely wrinkled striations; frons
smooth with delicate transverse striations centrally; vertex smooth, temples
smooth, except faint raised rugosities paralleling occipital carina posteriorly;
malar space smooth; occipital carina with several irregular short anteriorly
directed branches laterally (Fig. 5); ocellar triangle slightly raised, smooth;
eyes reduced, sparsely covered with short setae.

Thorax: Lateral area of prothorax deeply excavated, smooth and relative-
ly broad, delimited by well defined carinae (Fig. 5); mesonotum moderately
declivous, the lobes coarsely rugose with granular centers; notauli foveolate,
deeply impressed and slightly lighter than mesonotal lobes; scutellar furrow
deeply excavated, the anterior and posterior margins scalloped on either side
of a well developed median cross carina; scutellar disc rugose; mesopleural
disc smooth; sternaulus (Fig. 5) divided into two deep foveae by a strong median
cross carina, a well developed longitudinal carina extending posteriorly to
the mesocoxal cavity, forking about midway; prepectal carina complete; pre-
pectal area smooth with two weak horizontal carinae; fore tibia with 24 spines

Matthews: Nearctic Spathius 31

arranged in an irregular double row, a row of 5 spines at apex; hind tibiae with
raised parallel longitudinal ridges below, outer apical lobes with 3 spines; fore
and mid coxae smooth, hind coxae rugulose; hind tarsal segment 1 about 2. 4x
segment 2; second abscissa of radius distinctly longer than first intercubitus;
subdiscoideus arising above middle of brachial cell; propodeal carinae all well
developed, without trace of lateral branches (Fig. 4); basal carina about 1. 3x
the basal forks; segments of basal forks subequal on either side of costula in-
sertion; dorsal areas and areola with reticulate surface rugosity; area petio-
laris rectangular, about 3x as long as wide, subequal to the pentagonal areola
in length, the carina separating them strongly protuberant and visible as a
distinct toothlike projection in profile (Fig. 5); posterior-dorsal-lateral angles
very weakly protuberant.

Abdomen: Petiole arched and dorsoventrally thickened at spiracles (Fig. 5),
with smooth longitudinal striations dorsally; gaster smooth and shining, with
no trace of surface sculpture; lateral margins of tergite (2-3) not thickened
basally.

Color: Head shining orange brown, except face and malar area suffused with
yellow; maxillary and labial palpi stramineous; mandibles testaceous except
teeth darker; scape, pedicel and basal flagellar segments yellow brown, flagel-
lum becoming gradually light brown apically; thorax, propodeum, petiole and
gaster ferrugineous except prothorax which is testaceous; gaster shining, the
tergites with ill-defined subapical, subcuticular testaceous bands; legs yellow
brown except ferrugineous hind femora, tibiae not or indistinctly banded at base;
wings (Fig. 9) subhyaline, costa and basal third of stigma tan, other veins and
remainder of stigma medium brown; tegulae testaceous, shining; ovipositor
sheaths yellow brown basally, becoming gradually brown toward apex.

Variation: Paratype females. In addition to characters given in Table 1,
minor variations in number and development of the cross carinae of the scutel-
lar furrow, sternauli and area petiolaris occur, with up to 4 distinct scutellar
carinae and the sternauli either undivided or comprised of as many as 4 foveae.
The number of fore tibial spines varies from 17 to 26. The Maine paratype
differs from the others as follows: the sternauli are merely deep depressions
without thickened margins (hence not true foveae) and the carina extending from
the posterior margin to the mesocoxal cavity is totally effaced on the anterior
half; the prepectal carina is incomplete medially and there are additional short
horizontal carinae immediately above the prepectal area, which has in addition
a medial vertical carina; instead of delicate transverse striae, the frons has
delicate irregular raised rugosities medially; the petiole arches distinctly be-
yond the spiracles at about its geometric midpoint; the carina between the area
petiolaris and areola is not protuberant and the forks of the basal carina are
distinctly arched instead of straight. While these differences may prove to
merit specific designation, I chose to include it here until more material is
available.

Paratype males: The two known males agree with the holotype in color and
sculpture, except they have the frons smooth, the petiole more slender and
elongate, and one from Virginia lacks median cross carinae on the sternauli.

Paratypes: 19, same data as holotype [USNM]. Iowa: Mt. Pleasant, 19,
26 May 1934 [USNM]. Kansas: Onaga, 19, 2 June 1923 (Crevecoeur) [KSU].
Maine: Round Mt., 19, 16 July [MCZ]. Maryland: nr. Colesville, 19, 28
May 1914 (R. C. Shannon) [USNM]; Beltsville, 19, April 1941 (A. E. Davis)
[USNM]. Michigan: Ann Arbor, 19, 29 May 1962 (H. & M. Townes) [Townes];
St. Joseph Co., 19, 5 Aug. 1953 (R. R. Dreisbach) [Townes]; Wayne Co.,
Grosse Isle, 19, 13 Aug. 1956 (G. Steyskal) [Townes]. New York: Chappagua,

32 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

192, 18 May 1937 (L. L. Pechuman) [USNM]. Pennsylvania: Stoverdale, 19,
25 Aug. 1916 (W. S. Fisher) [USNM]. Tennessee: Chapin Sanctuary, East
Ridge, 1c’, 9 May 1952 (O. Peck) [CNC]. Virginia: Arlington, 1c’, 8 Sept.
1952 (K. V. Krombein) [USNM]. |

Remarks: This species may be distinguished from all other North Ameri-
can Spathius by the smooth mesopleural disc and the rugose scutellum and
mesonotal lobes. The unique form of the sternauli, lack of banded wings, and
propodeal profile (Fig. 5) are also useful diagnostic characters. Interestingly,
a very similar propodeal profile also occurs in S. cassidoris Nixon from New
Guinea (Nixon, 1943, p. 268). This distinctive species is named for its collec-
tor, Dr. Paul M. Marsh, in grateful appreciation for his continued interest
and encouragement throughout this study.

Distribution: The localities of the paratypes indicate that the range of
S. marshi is apparently eastern North America north of Virginia and Tennessee
and west to Iowa. In addition to the paratypes, I have seen specimens from
Illinois and Quebec, Canada.

Host: No hosts are known. The New York paratype is labeled from apple,
and that from Pennsylvania from hickory.

4, Spathius stigmatus, new species
(Figs. 6, 12, 13)

Holotype: - female, New York: Ithaca, 17 Aug. 1947 [CNC].

Description of type female: Body length, 4.4 mm; FWL, 2.9 mm; oviposi-
tor sheath length, 7.5 mm. Other measurements and ratios in Table 1.

Head: Antennae short, about equal to FWL; flagellomeres 18, subcubical,
those beyond the third less than 2x as long as wide; flagellomeres each with a
corona of 4 evenly spaced erect elongate setae about 4x as long as other flagel-
lar setae (Fig. 6); face transversely striate; frons centrally delicately trans-
versely striate, becoming faintly granular toward eyes and antennal bases;
vertex adjacent to ocellar triangle smooth, posteriorly becoming delicately
transversely striate, this continuing laterally onto temples to malar space;
mandibles striate basally below, the tips dark; ocellar triangle weakly raised,
appearing smooth; maxillary palpi slightly reduced, extending to about posterior
border of forecoxae, the segments laterally compressed and gently curved, the
apparent second longer than any other segment.

Thorax: Lateral area of prothorax narrow and poorly defined; propleura
horizontally carinate; mesonotum moderately declivous, the lobes evenly
granular; notauli deeply impressed anteriorly, converging to a broad, slightly
depressed longitudinally rugose area; scutellar furrow with about 10 evenly
spaced cross carinae; scutellum trapezoidal, about as long as basal width and
slightly convex, its disc uniformly granular, becoming weakly rugose posterior-
ly; mesopleural disc finely horizontally strigose; sternaulus shallow, crossed
by about 10 moderately developed carinae, these continuing posteriodorsally
onto mesopleural disc; prepectal carina complete; prepectal area with irregular
surface rugosities and about 4 to 6 short horizontal carinae arising along anteri-
or margin and extending posteriorly about to middle of area; propodeal carinae
basad of costulae obscured, entire dorsal area uniformly granular, becoming
rugose-reticulate apically; dorsal and lateral carinae well developed, as are
those enclosing the almost square area petiolaris; posterior-dorsal propodeal
angles weakly protuberant; tegulae delicately shagreened; fore tibia with 13

Matthews: Nearctic Spathius 33

Spines arranged in an irregular single row, a row of 6 spines at apex; outer
apical lobe of hind tibia with a row of 4 rather stout spines; hind coxae angled
basally below; hind tarsi shorter than hind tibiae, the basitarsi about 2.3x as
long as the second tarsomeres; second abscissa of radius about 1. 8x first
intercubitus; recurrens about equal to first abscissa of radius; subdiscoideus
nearly interstitial.

Abdomen: Petiole slightly arched at base, about equal in length to tergite
(2-3); petiole sculpture dorsally granular except finely longitudinally striate
on apical 1/4 with a pair of longitudinal dorsal carinae along each side; tergite
(2-3) smooth except for fine aciculation at extreme base, the lateral margin
weakly thickened at base; remainder of gaster appearing uniformly delicately
shagreened, strongest on epipleurites; ultimate tergite with fine basal longitu-
dinal striations (not visible unless tergites are distended).

Color: Head, thorax, propodeum, legs and petiole ferrugineous; gaster
castaneous, lighter at apex ventrally; maxillary and labial palpi light brown;
antennae light ferrugineous, the apical 6-8 flagellomeres becoming brown and
all flagellomeres with a narrow darker apical band; tibiae lightened on about
basal 1/4, appearing banded, this most apparent on hind tibiae; wings (Fig. 12)
infumated except subhyaline tips and a transverse band from the stigma base;
veins and stigma (except base) light brown; ovipositor sheaths ferrugineous,
the tips dark brown. 7

Variation: Paratype females. Gaster color in 2 females is completely light
ferrugineous. In others, tergite (2-3) is all or in part ferrugineous, the re-
mainder of the gaster being castaneous. The lighter specimens also show a
reduction in wing and leg band intensity. In about half of the paratypes, the
subdiscoideus is exactly interstitial. In sculpture, basal aciculation of tergite
(2-3) covers as much as the basal 1/3 of the tergite in some specimens and
carinae development on the prepectal area may be reduced or sometimes en-
tirely absent. Number of scutellar furrow carinae varies from 10 to 15, and
number of fore tibial spines from 12 to 15. Variation in measurements and
ratios is given in Table 1.

Paratype males: The five male paratypes agree with the females in all
important respects except the hind wings show distinct stigmal deveiopment
(Fig. 13) and the margin of tergite (2-3) is thickened for 2/3 of its length.

Paratypes: 299, 1c’, same data as holotype [CNC]. Delaware: New Castle
Co., 299, 5 & 7 June 1922 (F. M. Jones) [Townes]. I[llinois: Urbana, 12, 12
July 1920 (C. P. A.) [INHS]; Elsah, Jersey Co., Principia College, 19, 27
April 1943 (C. L. Remington) [USNM]. Maryland: Forest Glen, 19, 26 Apr.
1914 (O. Heidemann) [USNM]. Michigan: Livingston Co., G. Res., 19, 28
May 1944 (R. R. Dreisbach) [RDS]. New York: Syracuse, 299, .'"o", 21 Mar.
& 26 May 1916, N. Y. S. Coll. For. Lot Nos. H295, H438 (M. W. Blackman
and H. H. Stage) [Syracuse, Hamden]. North Carolina: Tryon, 399, Hopk.
No. 3605c and 1464f (W. F. Fiske) [USNM]. Ohio: Wayne Co., 19, 2 July
1941 [OSU]. Pennsylvania: Hopk. No. 10383 (Linglestown), 799, 25 June 1914
(H. B. Kirk) [USNM, MCZ]. Texas: Hopk. No. 3861 (Montell), 1c’, April 1907
[USNM]. Virginia: E. Falls Church, 19, 9 May 1909 (H. L. Viereck) [USNM].
Wisconsin: Gay Mills, 19, 8 July 1953 (C. L. Fluke) [RDS]. Canada: Hull
(Quebec), 19, 19 Aug. 1894, 1o’, 23 June 1895 [CNC]; Ontario, Ottawa, 19
[USNM].

Remarks: The very long ovipositor, shagreened gaster and form of the
flagellomeres (Fig. 6) readily distinguish the females. Presence of a stigma
in the male hind wing (Fig. 13) will separate the males from all other known
North American Spathius, and is the basis for the specific name, stigmatus.

34

Table 1.

brachyurus species group.

Contrib. Amer. Ent. Inst.,

vol. 4, no. 5, 1970

number below is for holotype or lectotype.

n

number of
flagellomeres

body length
(mm)
FWL (mm)
MS/EH
TW/EH
IOD/EH
OTW/OED
OED/OOD
Pet. L/(2-3)L

ICu;/Ro9

OSL/FWL

brachyurus

25

19-25
broken

0,67-0. 82
0.67

1.03-1.29

In Ashmead’s (1893) key, S
vancher, and indeed one specimen from CNC bears this determination label
in Ashmead’ s handwriting.

Distribution: Paratype localities suggest a distribution covering eastern
North America, although except for Texas the species has not been recorded

south of North Carolina. Additional specimens add only Indiana.

brevipalpus

0. 73-1. 00
0.95

1.00-1. 30
1.26

1V08+3.:90
1.90

0.53-0. 65
0.56

0.82-0. 83
0.82

1.04-1.07
1.04

0.58-0. 67
0.58

0.52-0. 55
0.55

marshi
13

0. 78-1. 00
1.00

1,.0=2.55
1.17

1.50-1. 89
1.89

0.53-0. 72
0.56

0.82-0. 95
0.90

1.00-1. 50
1,25

0. 77-0. 96
0.78

0. 69-0. 96
0.73

Measurements and ratios of selected characters for females of the
Top line in each instance gives range;

stigmatus
13

2.4-3.
2.9

0. 70-0, 96
0.88

1.04-1. 23
1.08

1.33-1.58
1.36

0. 56-0. 69
0.59

0. 67-0. 80
0.71

0.90-1.22
1.00

0.50-0. 80
0.56

2.20-2. 70
2.60

. Stigmatus s keys out exactly to S. laflammei Pro-

Hosts: Hopkins cards in the U. S. National Museum and records of Black-
man and Stage (Syracuse) indicate that S. stigmatus has been reared from
hickory dead 3-4 years in association with the cerambycid Neichnea laticornis

Say and the clerid Phyllobaenus dislocatus Say (from Hopk. No. 10383).

Black-

Matthews: Nearctic Spathius 35

man and Stage’s (1924, p. 165) reference to Spathius trifasciatus Riley is prob-
ably a misidentification and instead almost surely refers to this new species,
since the data on the specimens is in almost exact agreement. They conclude
that one of its hosts is Magdalis olyra Herbst. (See also ‘‘Hosts’’ and
‘‘Remarks’’ under S. trifasciatus.) The paratype from Texas was reared

from pecan infested by cerambycids which had previously emerged; several
associated insects are listed under Hopkins No. 3861. Of two additional females
in the U. S. National Museum, one was reared from Celtis (Hopk. No. 2573)
and the other bears the label ‘‘parasitic on Xyloterinus politus [Scolytidae] in
maple. ’’

5. Spathius calligaster, new species
(Figs. 17, 20)

Holotype. - female, South Carolina: Georgetown, 28 Mar. 1938 (F. C.
Craighead) ex. Melasis pectinicornis ? in Myrtle [USNM Type No. 70500].

Description of type female: Body length, 5.6 mm; FWL, 3.7 mm; oviposi-
tor length, 5.5 mm.* Other measurements and ratios in Table 2.

Head: Head slightly dorsoventrally compressed (Fig. 17), the eyes slanted
obliquely, creating a relatively broader lower temple area; flagellomeres 37;
face transversely strigose-granular; frons transversely striate, the center
area Slightly rugose; vertex delicately transversely striate anteriorly, posteri-
orly becoming smooth in front of occipital carina; temples delicately strigose
more or less parallel to occipital carina, this strongest posteriorly, becoming
Smooth at eye and malar space; ocellar triangle very weakly raised and deli-
cately granular, areas between the lateral ocelli and eyes, separating vertex
and frons, smooth.

Thorax: Thorax strongly dorsoventrally compressed (Fig. 17), about 3x
as long as high at tegulae; lateral areas of prothorax distinct and moderately
excavated, the anterior margins converging dorsally to form a distinct trans-
verse swelling slightly in front of the anterior mesonotal margin (Fig. 17);
notauli crenulate anteriorly, meeting in a broad slightly depressed rugose
area; mesonotum sloping gradually to pronotum, the lobes evenly granular;
scutellar furrow shallow, narrowest medially, about as wide as distance be-
tween lateral ocelli and crossed by about 7 variously developed carinae, some
partially obscured; scutellum nearly flat and evenly granular on disc; meso-
pleural disc nearly smooth immediately above sternaulus, becoming delicately
horizontally strigose to reticulate-rugose on upper portions and granular pos-
teriorly; sternaulus moderately excavated, a gently curved furrow, deepest
about midway between fore and middle coxae and with several irregular raised
cross carinae which do not continue onto mesopleural disc; prepectal carina
incomplete dorsally above sternaulus (Fig. 17); prepectal area smooth and
somewhat reflexed posteriorly, with a few hairs and faint irregular raised
rugosities; propodeum nearly horizontal; propodeal carinae mostly obscured,
lateral carinae best developed; areola, costulae and basal forks completely
obscured by reticulate rugosity which merges into the coarsely granular dor-
sal areas; basal carina faintly developed, rather long, extending to about half

*
In this species the ovipositor sheaths tend to coil strongly on all dried specimens,
hence the actual ovipositor length was measured.

36

Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

the length of propodeum; fore tibia with 19 spines arranged in an irregular
double row and with a row of 7 spines at apex; outer apical lobe of hind tibia
with no spines present; hind coxae elongate, slender and coarsely granular all
over, except for very faint oblique striation dorsally; outer face of hind tibiae
irregularly longitudinally striate; tegulae granular medially, the outer portions
smooth; subdiscoideus vein exactly interstitial; first abscissa of radius dis-
tinctly shorter than recurrens; first intercubitus with a distinct curvature

(Fig. 20); mediella with 3 distinct abscissae.

Table 2. Measurements and ratios of selected characters for females of the

calligaster and longipetiolatus species groups.

Top line in each

instance gives range; number below is for holotype or lectotype;
longipetiolatus is known only from the type.

n

number of
flagellomeres

body length

(mm)
FWL(mm)
MS/EH
TW/EH
IOD/EH
OTW/OED
OED/OOD
Pet. L/(2-3)L
ICu, f Ro

OSL/FWL

calligaster longipetiolatus
(holotype male)

elegans

0.49-0. 65

0.49

evansi

11

35-48
45

3. 1-8.5
6.4

Matthews: Nearctic Spathius Ot

Abdomen: Petiole straight, not arched basally, sculpture coarsely granu-
lar to rugose reticulate with a pair of wrinkled longitudinal carinae along most
of each side above the spiracles; tergite (2-3) entirely coarsely granular-
reticular except apical margin which is smooth, the sculpture strongest on basal
half, this delimited by a faint transverse line at about midpoint of tergite; ter-
gite (2-3) thickened laterally almost entire length; remaining tergites granular
basally, the apical margins distad of the setae rows smooth; intensity of sculp-
ture progressively diminishes on each succeeding tergite, becoming punctulate
on penultimate tergite; epipleurites lacking obvious sculpture, at most dull
shagreened.

Color: Body color more or less uniformly pale testaceous except the
posterior margin of the mesonotum and scutellar carinae are dark brown and
gaster with very faint darker subcuticular transverse bands dorsally and pale
stramineous ventrally; antennae, tegulae and fore legs honey yellow; maxillary
and labial palpi, all trochanters, basal 1/4 to 1/3 of tibiae, and fore and mid
coxae cream white; hind coxae, mid and hind femora and tibiae darker, dull
testaceous; mandible tips ferrugineous; ovipositor sheaths pale testaceous,
the tips darker, almost brown; wings banded as in Figure 20; costal vein honey
yellow, other veins tan, the basal vein and media darkest; basal 1/3 of stigma
pale stramineous.

Variation: Paratype females. In addition to variation indicated in Table 2,
body color may be generally darker to more or less uniformly dull ferrugineous;
however, the basic pattern prevails, i.e., the fore and mid coxae are always
much lighter than the hind coxae, and the antennae, fore legs and tegulae are
usually at least slightly lighter than the predominating body color; flagello-
meres, at least basal ones, often with a darker narrow apical rim; eyes gray
to black, very rarely showing any iridescence in white light; prepectal carinae
variously developed, almost completely obscured in the smallest female and
occaSionally appearing complete in larger individuals; subdiscoideous position
somewhat variable, not always interstitial; gastral sculpture beyond tergite 4
varying in degree of development, usually much diminished in intensity, the
apical tergites often nearly smooth; number of fore tibial spines varying from
14-22, with a row of 6 or 7 at apex.

Paratype males: In general variation in the males parallels that found in
females. Additionally, males often have some tergites beyond (2-3) laterally
thickened, and also the hind wings show a slight tendency towards stigma
development.

Paratypes: 1299, To‘o’, same data as holotype [USNM]. Canada, Quebec:
La Trappe, 399, 1c’, 20 July 1947, 27 July 1948, 12 July 1950, 3 Aug. 1943
(J. Ouellet) [Montreal, CNC, MCZ]. U.S.A., Illinois: White Heath, 299,

2 Aug. 1940 (Ross & Riegel) [INHS]. Maryland: Patuxent Wildlife Ref., 10
(P. M. Marsh) [USNM]. New Jersey: Metuchen, 1c’, 13 June 1964 (Porter)
[Porter]; Moorestown, 1c’, 19, 23 June and 13 Aug. 1939 (H. & M. Townes)
[Townes]. North Carolina: Tryon, 19, 10 May 1946 (M. Townes) [Townes].
Virginia: Plummers Id., 1c’, 19, 6 June 1916, reared from Carpinus carolini-
ana, Hopkins No. 14023 (H. S. Barber & T. E. Snyder) [USNM].

~ Remarks: In addition to its somewhat dorsoventrally compressed body
(Fig. 17), this species may be distinguished from all other North American
Spathius by the combination of the coarsely sculptured gaster, especially ter-
gites (2-3) and 4, the thickened lateral margin of tergite (2-3) for most of its
entire length, the absence of spines on the outer apical lobe of the hind tibia,
the dorsal transverse swelling on the prothorax and the first abscissa of the
radius distinctly shorter than the recurrent vein. In addition, the petiole is

38 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

straight and the subdiscoideus vein is usually more or less interstitial. In its
characters this species resembles those of Nixon’s labdacus group.

Distribution: The known range corresponds with the paratype localities
which are scattered through the eastern United States and Quebec, Canada,
including Illinois, Maryland, New Jersey, North Carolina, South Carolina and
Virginia. |

Hosts: The holotype series from Georgetown, South Carolina, was reared
with the eucnemid, Melasis pectinicornis Melsh. in myrtle, according to their
labels and data filed under Hopkins No. 17412e in the U. S. National Museum.
No other associated insects are mentioned. The paratypes from Virginia were
reared from Carpinus caroliniana in association with Leptura mutabilis Newm.,
Leptostylus macula Say (both Cerambycidae) and Melasis pectinicornis, accord-
ing to data filed under Hopkins No. 14023 in the U. S. National Museum.

6. Spathius longipetiolatus Ashmead
(Fig. 22)

Spathius longipetiolatus Ashmead 1893, Can. Ent. 25: 70.
[Type locality: Florida: Jacksonville (Lectotype male in USNM,
Type No. 52647)].

Redescription of lectotype male: Body length, 3.4 mm; FWL, 2.4 mm;
other measurements and ratios in Table 2.

Head: Flagellomeres 22; face transversely strigose; frons with several
polished transverse striations; vertex strongly transversely striate like frons,
those of vertex separated from those of frons by a smooth transverse space
about width of ocellar triangle, the striae nearly obscured medially; temples
smooth and shining; ocellar triangle slightly raised, polished.

Thorax: Lateral areas of prothorax broad, deeply excavated and polished,
bordered by a well defined carina and crossed by 7-8 oblique carinae; notauli
shallowly crenulate anteriorly, meeting in an area of weak rugosity from which
emerge 2 posterior carinae extending to posterior margin of mesonotum; meso-
notum abruptly declivous, the lobes evenly granular; scutellar furrow with 3
prominent cross carinae flanked laterally by a less strong carina on each side;
scutellum slightly projecting and more or less smooth, with faint sculpture
laterally and posteriorly; mesopleural disc smooth and shining above sternaulus,
becoming somewhat rugose dorsally; sternaulus deeply excavated, especially
posteriorly, smooth, crossed by 6 or 7 carinae, the space posterior to furrow
with curved striae extending to mesocoxal cavity; prepectal carina complete;
prepectal area smooth with slight surface irregularity; propodeal carinae all
distinct; basal carina about 1/4x as long as its forks; fore tibia with at least 12
spines arranged in a single row; outer apical lobe of hind tibia with apparently 4
spines; tegulae smooth; subdiscoideus vein nearly interstitial; recurrent vein
almost 2x first abscissa of radius; first intercubitus and second abscissa of
radius subequal.

Abdomen: Petiole slender and elongate, about 2/3 as long as thorax and
irregularly longitudinally strigose dorsally; tergite (2-3) nearly smooth and
shining, very faintly shagreened at base; remaining tergites smooth and shining,
the lateral margins thickened; genitalia as in Figure 22.

Color: Body color light ferrugineous, the head lighter testaceous and
petiole darker, almost castaneous; antennae testaceous on basal segments,
becoming gradually darker apically; legs unicolorous, honey yellow, the tibiae

Matthews: Nearctic Spathius 39

not banded and the tarsi very slightly darker; maxillary and labial palpi and
tegulae testaceous; propodeal spiracle testaceous, contrasting with ferrugineous
propodeum; wings subhyaline, the veins light yellowish-brown; stigma wholly
brown, very slightly lighter basally.

Variation: Paralectotype male. Agrees with type in all apparent respects,
but its poor condition makes adequate comparison impossible.

Female: Unknown.

Remarks: Since its description, no additional collections or records of this
Species have been made, and the two males on which this species is based re-
main unique among the Nearctic species. The combination of the smooth scu-
tellum, slender elongated petiole, strong transverse striation on the vertex and
deeply excavated sternauli and lateral areas of the prothorax make this a very
distinctive species and I found no specimens which possessed these features
among the material studied. The elongate petiole most resembles that of
Spathius elegans, sp. n., but thetwo differ strikingly in most other features.

As was true for several of the Spathius described in his 1893 paper, Ash-
mead gave no formal description of S. longipetiolatus beyond the characters
used in his ‘‘Table of Species’’. Although not specifically stated by Ashmead,
there seems little doubt that he had at least two males before him, since he
states a range for the length of ‘‘2.5-3.0 mm’’. Inthe U. S. National Museum
collection are two males bearing three identical labels: ‘‘Jacksonville, Fla.’’,
‘‘Spathius longipetiolatus Ashm.’’ in Ashmead’s handwriting, and red ‘‘USNM
Type No. 52647’’ on one and ‘‘USNM Paratype No. 52647’? on the other. Both
Specimens were very dirty and embedded in glue; one I cleaned, extracted the
genitalia, remounted, and have designated as lectotype. The other (missing
the wings and abdomen) I labeled as paralectotype without attempting to clean
Hou

Distribution: Known only from the type locality, Jacksonville, Florida.

Host: Unknown.

7. Spathius elegans, new species
(Fig. 22)

Holotype.- female, Maryland: Plummers Id., vi-6-13, reared Acer, Hopkins
No. 11338 (H. S. Barber) [USNM Type No. 70501. ]

Description of type female: Body length, 5.6 mm; FWL, 4.2 mm; oviposi-
tor sheath length, 6.0 mm. Other measurements and ratios in Table 2.

Head: Flagellomeres 31; face finely transversely strigose-granular,
becoming granular at eyes; frons strongly transversely striate, this ending
abruptly at level of lateral ocelli; vertex smooth and shining, no trace of any
sculpture; temples mostly smooth and shining, a very faint trace of delicate
vertical strigosity next to occiput, but perfectly smooth at eye and malar
space; ocellar triangle weakly raised with weak central sculpture.

Thorax: Lateral areas of prothorax relatively broad and well defined
anteriorly with prominent diagonal cross carinae, posteriorly becoming ob-
scured in the irregular granular rugosity which covers most of propleura;
pronotum with a well developed transverse ridge just behind the anterior mar-
gin; notauli crenulate anteriorly, converging into a broad moderately depressed
rugose area; mesonotum moderately declivous, the lobes more or less evenly
granular; scutellar furrow deep, with about 4 well developed cross carinae;
scutellum very slightly convex and projecting, the disc granular, becoming

40 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

weakly rugose posteriorly; mesopleural disc irregularly rugose to granular;
sternaulus a distinct furrow extending about 2/3 of mesopleuron, with several
irregular to almost rugose cross-carinae; prepectal carina complete, well
developed; prepectal area smooth, shining, with a weak horizontal carina just
above the fore coxa; propodeal carinae all visible, tending to be somewhat
irregular and wrinkled; dorsal areas coarsely granular to weakly rugose; basal
carina quite well developed, straight and about equal in length to its less well
developed forks; areola and area petiolaris rather narrow and elongate, the
sides subparallel, weakly rugose basally, the rugosity stronger apically; fore
tibia with 39 spines arranged in row of 2-3 with an irregular row of 9 at apex;
outer apical lobe of hind tibia with a row of 4 spines; hind basitarsus distinctly
longer than second tarsomere (ratio of 32/13) and tarsomere 3 slightly shorter
than tarsomere 5; subdiscoideus not interstitial; second abscissa of radius
slightly shorter than first intercubitus; mediella with 3 abscissae.

Abdomen: Petiole slender and nearly straight, about 3/4 as long as thorax
(Fig. 22), granular to weakly rugose dorsally, becoming irregularly longitudi-
nally striate apically; tergite (2-3) with basal 3/8 prominently longitudinally
strigose-granular, abruptly delimited apically, followed by equal area of
delicate scaly-reticulation grading imperceptibly into apical 1/4 which is
smooth and shining; lateral margin of tergite (2-3) thickened on about basal
3/8; tergite 4 with faint sparse punctulation anterior to the setae row; remain-
ing tergites apparently smooth.

Color: Head, pro- and mesothorax testaceous; propodeum and petiole honey
yellow, distinctly lighter than remainder of thorax or gaster; gaster shining
ferrugineous except for a honey yellow transverse band on apical 1/4 of tergite
(2-3); legs more or less concolorous with head and thorax, except for stramine-
ous trochanters, fore and mid coxae and basal 1/4 to 1/5 of tibiae; palpi also
stramineous; antennae honey yellow basally becoming progressively darker
testaceous towards apex; fore wing distinctly banded, the basal third of stigma
whitish; veins testaceous to brown; eyes dark gray.

Variation: Paratype females. Considerable variation in size, color and
intensity of sculpture occurs; color is most commonly dull ferrugineous, but
varies from testaceous to mostly piceous; however, the pattern always remains,
especially having the propodeum and petiole distinctly lighter than the remainder
of the thorax and gaster, even to the naked eye; the lighter transverse band on
the apical margin of tergite (2-3) is usually discernible on the majority of
individuals, but may occasionally be obscured, the gaster appearing unicolorous;
eyes never with iridescent hue in white light; intensity and development of the
striation on the frons varies greatly, smaller examples tending to have almost
smooth frons; the vertex, however, is always smooth and shining; sculpture on
the basal 3/8 of tergite (2-3) is always distinctly present, but that on the next
3/8 is variable, sometimes almost completely obscured; the rugose sculpture
of the mesopleuron is also variously expressed, occasionally obscured; scu-
tellar furrow with 3-8 variously developed cross carinae, some of which may
be wrinkled and irregular medially; relative lengths of the first intercubitus
and second abscissa of the radius are not reliable, about half of the individuals
having the first intercubitus shorter than or equal to the second abscissa of
radius; hind basitarsus is always distinctly longer than the second tarsomere
and the fifth tarsomere is usually slightly longer than the third; number of
fore tibial spines varies from 22-48 with a row of 7-10 at apex; outer apical
lobes of hind tibiae with a row of 3-5 spines. Variation in measurements and
ratios given in Table 2.

Paratype males: In general, variation tends to parallel that in females.

Matthews: Nearctic Spathius 41

Smaller males are very difficult to distinguish because the useful sculpture
differences tend to become obscured.

Paratypes: 629, same data as holotype [USNM]. Canada, Ontario: Pt.
Pelee, 19, 1 July 1957 (H. & M. Townes) [Townes]; Ottawa, 19, 16 Aug. 1947
(W. R. M. Mason) [CNC]. Quebec: Messines, 19, 10 July 1947 (W. R. M.
Mason) [CNC]; Wakefield, 19, 20 June 1946 (G. S. Walley) [CNC]; Gatineau
Park, Old Chelsea, 399, 8 July 1966 (D. R. Smith) [USNM]. U.S.A., Alabama:
Chambers Co., Langdale, 19 (H. H. Smith) [USNM]. Connecticut: Redding,
19, 28 July 1933 (Melander) [MCZ]. Florida: Gainesville, 19, 19 May 1918
(C. J. Drake) [Syracuse]. [linois: Urbana, U. of Ill. Woods, 299, 12 July
and 22 Aug. 1937 (Sarah Jones) [INHS]; Jonesboro, 19, 21 June 1939 (Burks &
Riegel) [INHS]; Warren, 19, 13 June 1955 (R. L. Fischer) [MSU]. Kansas:
Douglas Co., 192, 30 May 1951 (J. G. Rozen) [Berkeley]. Kentucky: Golden
Pond, 19, 10 June-10 Aug. 1964 (S. G. Breeland) [Townes]. Maryland: Takoma
Pk., lo’, 299, 5 July 1942, 9 May & 10 July 1943 (G. F. Townes and H. & M.
Townes) [Townes]; Bowie, 2c‘o", 24 June 1945 (H. & M. Townes) [Townes].
Massachusetts: Sharon, 19, 3 Aug. 1904 [MCZ]; Petersham, 19, 17 July [MCZ].
Michigan: Midland Co., 19, 7 Sept. 1944 (R. R. Dreisbach) [USNM]. New
Hampshire: Mt. Madison, 192, 24 June 1938 (H. & M. Townes) [Townes]; White
Mts., Dolly Copp, 299, 13 July 1931 (A. L. Melander) [MCZ]. New Jersey:
Metuchen, Woodwild Pk., 399, 6 Sept. 1955, 15 Aug. 1956, 8 July 1955
[Porter]. New York: Bemus Pt., 599, 2c’o’, 26 June and 6, 8, 11, 22, 31
July 1937 (H. K. Townes) [Townes]; Millwood, 399, 1c", 21 June 1936 (H. K.
Townes) [Townes]; Colden, 19, 7 June 1908 (M. C. Van Duzee) [CAS]; Albany
Co. nr. Rensselaerville, Huyck Preserve, 299, 28 June and 1 July 1967 (R. &
J. Matthews) [MCZ]. North Carolina: Pisgah Mt., 19, 21 June 1940, 4800-
5300 feet (H. & M. Townes) [Townes]; Crabtree Mds., Yancey Co., 19, 21 Aug.
1950, 3600 ft. (H., M. & D. Townes) [Townes]. Pennsylvania: Wilawanna, 299,
25 July 1938 (R. H. Crandall) [Arizona]. Tennessee: Green Briar Cove, Mount
trail, GSMN Park, 19, 12 June 1946[RDS]. Vermont: Jacksonville, Laurel
Lake, 19, 15 July 1939 (H. D. Pratt) [Minnesota]. District of Columbia:
Washington, 399, 26 & 29 Aug. 1907, 7 Sept. 1907, bred from Carpinus ameri-
canus (J. A. Hyslop) [USNM].

Remarks: This is one of the most frequently collected Spathius species al-
though it has so far only rarely been reared. I have seen in excess of 200
Specimens. A highly variable species, it may be readily identified by the com-
bination of the smooth, shining vertex, rugose mesopleura, long hind basitarsus
(always greater than 2x the second tarsomere) and the distinctly lighter colored
propodeum and petiole. A number of strongly dorsoventrally compressed in-
dividuals of S. elegans were also noted; these are sometimes So greatly dis-
torted that certain identification is difficult.

Distribution: Throughout eastern North America. In addition to the para-
types, I have seen specimens from Georgia, Iowa, Maine, Mississippi, Nebras-
ka, Rhode Island, South Carolina, Virginia, West Virginia and Wisconsin. I
also tentatively include here a single female from San Juan Island, Washington
(RDS).

Hosts: No positive host associations are known. The holotype series was
reared from maple under Hopkins No. 11338, but no host associations were
made. Two males from Plummers Island, Maryland, in the U. S. National
Museum were reared from Acer under Hopkins Nos. 10178 and 11341; another
male was reared from black walnut, according to Hopkins No. 2566. The
paratypes from Washington, D. C., were reared from Carpinus (blue beech)
and another individual I saw was reared from hickory. A teneral male, probably

42 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

of this species, was reared from powderposted wood, probably as a parasite
of the anobiid Hadrobregmus, according to Hopkins No. 34373c. Brues’ (1927)
record of Spathius simillimus is a misidentification and should be changed to
this species, as his single female reared July 17 is in the MCZ (paratype from
Petersham, Mass.).

8. Spathius evansi, new species
(Fig. 21)

Holotype.- female, Wyoming: Moran, Jackson Hole Bio. Sta., Aug. 1-14,
1961, 6750 feet elev. (H. E. Evans) [MCZ Type No. 31700].

Description of type female: Body length, 6.4 mm; FWL, 4.7 mm; oviposi-
tor sheath length, 9.5 mm.

Head: Flagellomeres 45; face transversely strigose granular; frons strongly
transversely striate, becoming delicately granular at eyes; vertex very faintly
transversely strigose to scaly reticulate anteriorly becoming smooth posteri-
orly at occiput; temples posteriorly strigose more or less parallel to occipital
carina, quickly diminishing in intensity towards eye, malar space and vertex;
ocellar triangle slightly raised and appearing granular.

Thorax: Lateral areas of prothorax moderately developed, smooth, crossed
by several diagonal carinae, these becoming obscured posteriorly in the granu-
lar to horizontally strigose sculpture of remainder of propleura; notauli deep
and crenulate anteriorly, meeting in a broad strongly depressed rugose area,
dominated by two wrinkled longitudinal carinae which end at posterior margin
of mesonotum; mesonotum moderately declivous, the lobes evenly granular,
the median lobe with a weak longitudinal depression medially; scutellar furrow
relatively broad, wider than the space between the lateral ocelli, and crossed
by 9 more or less parallel carinae of which the center one is most strongly
developed; scutellum slightly raised and projecting posteriorly, the disc evenly
granular; mesopleural disc coarsely granular to granular-rugose dorsally;
sternaulus a relatively broad, distinct furrow about 2/3 length of mesopleuron
and crossed by about 13 more or less parallel vertical carinae; prepectal cari-
na strongly developed, especially ventrally, but almost obscured just above
sternaulus; prepectal area faintly scaly-reticulate; mesosternum evenly granu-
lar, except rugose midventral line; propodeal carinae tending to be obscured
except those surrounding the area petiolaris and the lateral carinae; basal cari-
na irregularly wrinkled, about 1.2x its forks which are also irregularly wrin-
kled; costulae nearly obscured by weak rugosity; dorsal areas granular; areola
with weak granular rugosity over basal half, apically with 3 weak short longi-
tudinal carinae; area petiolaris smooth, about as long as broad; fore tibia with
25 spines arranged in an irregular double row and a row of 6 spines at apex;
outer apical lobe of hind tibia with 2 or 3 inconspicuous spines; hind coxae
elongate and uniformly scaly-reticulate to finely granular with no evidence of
oblique striations dorsally; hind tarsomere 3 about 1.4x as long as tarsomere 5;
tegulae with inner halves granular, outer halves smooth; subdiscoideus vein
not interstitial.

Abdomen: Petiole straight, slender and nearly as long as thorax, coarsely
granular dorsally with irregular longitudinal striation apically; tergite (2-3)
with prominent sculpture dorsally on basal 4/5 as follows - basal 1/2 longitu-
dinally granular-strigose grading into a granular to scaly-reticulate area of
less intensity, the apical 1/5 smooth and shining; lateral margin of (2-3) thick-

Matthews: Nearctic Spathius 43

ened on basal 1/2; remaining tergites apparently smooth except very weak
_ punctulation visible laterally on tergites 5, 6, and 7 anterior to setal row;
epipleurites dully shagreened.

Color: Body color more or less uniformly dull ferrugineous, except
sculptured part of tergite (2-3) and basal 1/2 of tergite 4 are suffused with
golden brown while the remaining gastral tergites approach castaneous; legs
concolorous with body except testaceous trochanters, fore and mid coxae, and
whitish-banded basal 1/6 to 1/4 of tibiae; antennae and ovipositor sheaths
testaceous, both becoming brown apically; tegulae, maxillary and labial palpi
testaceous; wings distinctly banded as in Figure 21; veins brown; basal 1/3 of
stigma pale whitish, remainder brown.

Variation: Paratype females. Agree with holotype in most features, al-
though body color is commonly lighter, dark testaceous, especially in individu-
als from eastern North America; propodeum, petiole and gaster visibly lighter
than remainder of thorax in some individuals, and gaster often of one hue in-
stead of darker apically as in holotype; paratype from New Mexico with the
median mesonotal lobe suffused with golden brown; eyes usually a dull greyish,
never with any iridescent hue in white light; vertex sculpture sometimes ob-
scured, but sculpture on tergite (2-3) always developed; number of fore tibial
Spines varies from 25-36 with a row of from 5-9 spines at apex; outer apical
lobe of hind tibia with 2-5 spines; variation in measurements and ratios given
in Table 2.

Paratype males: In general, variation tends to parallel that in the females.

Paratypes: Canada, Quebec: La Trappe, 299, 10 July and 2 Aug. 1943
(J. Ouellet) [Montreal, CNC]. U. S. A., Florida: Jackson Co., Fla. Caverns
St. Pk., 19, 23 June 1961 (H. V. Weems, Jr.) [USNM]. Iowa: Ames, 19, 4
Aug. 1951 (W. S. Craig) [USNM]. Maryland: Patuxent Ref., Bowie, 399, 30°C",
9 Aug. 1950 (R. T. Mitchell) [USNM]; Plummers I., 20‘o", 28 May and 4 July
1911 (J. C. Crawford) [USNM]. New Jersey: Metuchen, Woodwild Park, 599,
21 Aug. 1953 (377), 10 Aug. 1953 (501), 5 Sept. 1954 (1153), 12 June 1955 (1258)
(C. C. Porter) [Porter, ao. New Mexico: Cimarron, 19, 23 Aug. 1955,
9500 ft. (D. Townes) [Townes]. New York: Bemus Pt. 19, 25 July 1937
(H. K. Townes) [Townes]. North Carolina: Tryon, 19, Hopk. U. S. 1577e,
Castanea dentata (W. F. Fiske) [USNM]. Virginia: Great Falls, 19, 21 July
1962 (G. Steyskal) [USNM]. West Virginia: Hardy Co., Lost River St. Pk.,
19, 2o%o", 1-14 Aug. 1960 (K. V. Krombein) [USNM].

Remarks: This species is the largest of the North American Spathius. It
may be distinguished by the combination of the long slender straight petiole,
very long ovipositor (about 1.6-2.0x FWL), tergite (2-3) sculptured on about
basal .75-.80, strongly granular mesopleural disc and mesosternum, and hind
tarsomere 3 distinctly longer than tarsomere 5. It is named for Dr. Howard
E. Evans in grateful appreciation for his interest and guidance during my
Harvard years.

Distribution: Apparently widespread throughout all but the western coast
of North America. In addition to the paratype localities I have seen specimens
from Colorado, Ohio, Maine, Michigan, Pennsylvania and South Carolina.

Hosts: No hosts are known. The paratype from North Carolina was taken
from under the bark of chestnut, Castanea dentata, according to Hopkins No.
1577Te.

44 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

9. Spathius sequoiae Ashmead
(Figs. 16, 27)

Spathius sequoiae Ashmead, 1889 (1888), Proc. U. S. natn. Mus. 11: 625.
[Type locality: California: Alameda Co. (lectotype female in USNM,
Type No. 2925)].

Spathius californicus Ashmead, 1893, Can. Ent. 25: 71. [Type locality:
California: Santa Cruz Mts. (holotype male in USNM, Type No. 14660)].
New synonymy.

Spathius brunneri Viereck, 1912, Proc. U. S. natn. Mus. 42: 627-628. [Type
locality: Montana: Columbia Falls (holotype female in USNM, Type No.
14730)]. New synonymy.

Females: Body length, 2.5-4.1 mm; FWL, 2.1-3.6 mm; ovipositor sheath
length, 0.8-2.8 mm. Other measurements and ratios in Table 3. |

Head: Flagellomeres 22-32; face transversely strigose-granular and some-
what shining; frons delicately transversely striate to almost smooth and shining;
vertex virtually smooth and shining to transversely strigose; temples apparently
smooth but usually with faint wrinkled sculpture posteriorly adjacent to occipital
carina; ocellar triangle weakly raised, smooth to weakly strigose granular.

Thorax: Lateral areas of prothorax well developed and extending to pro-
coxal cavities, with several irregular oblique cross carinae; remainder of
propleura longitudinally strigose to weakly rugose; notauli distinct, crenulate
anteriorly, meeting in a broad shallow area of weak rugosity from which often
emerge two more or less parallel longitudinal carinae which extend to the
posterior margin of mesoscutum; mesonotum abruptly declivous, the lobes
evenly granular, the middle lobe sometimes with a faint longitudinal depres-
Sion medially; scutellar furrow variable, crossed by 3-9 variously developed
carinae, some of which are often wrinkled or partially obscure; scutellum
slightly projecting, the disc delicately granular to scaly-reticulate and lateral
faces irregularly obliquely striate; mesopleural disc highly variable, typically
delicately striate, many originating as weak extensions of sternaulus cross
carinae and converging posteriorly at speculum, but occasionally almost smooth
above sternaulus or weakly scaly-reticulate to delicately strigose-granular;
sternaulus relatively broad and well impressed, crossed by several delicate
closely spaced cross carinae, the more posterior elements stronger, several
curving posteriorly and continuing to mesocoxal cavities, the space following
the sternaulus distinctly obliquely striate; prepectal carina complete, pre-
pectal area smooth to irregularly surfaced, with sparse setae; propodeal cari-
nae all sharply defined and distinct; dorsal areas granular to weakly rugose;
relative lengths of basal carina and its forks variable; costulae approximately
perpendicular to basal forks and directed slightly anteriorly; areola typically
an elongate, almost parallel-sided trapezoid enclosing irregular raised sculp-
ture, occasionally broadened to almost diamond-shaped; area petiolaris rec-
tangular to sub-triangular, never longer than 2/3 of areola and usually devoid
of sculpture; fore tibia with 13-33 spines arranged in double to occasionally
triple irregular rows and a row of 5-8 spines at apex; outer apical lobe of
hind tibia with 2-5 spines; hind coxae weakly obliquely striate dorsally; hind
tarsomere 3 about equal to tarsomere 5; subdiscoideus not interstitial; first
abscissa of radius shorter than recurrent vein, so that second abscissa of
radius and second abscissa of cubitus converge Slightly apically.

Abdomen: Petiole granular to irregularly longitudinally strigose apically

Matthews: Nearctic Spathius

45

and arched at base; tergite (2-3) varying from apparently smooth or very

faintly sculptured at extreme base, to a distinct area of weak scaly-reticulate

?

aciculate or granular sculpture on at most about basal 3/8; remaining tergites
smooth, epipleurites nearly always faintly sparsely punctuate; tergite (2-3)
weakly thickened laterally at base; ovipositor slightly curved, typically 0. 6-

0.8x FWL but rarely shorter.

Table 3. Measurements and ratios of selected characters for females of the
sequoiae species group. Top line in each instance gives range;

number below is for holotype or lectotype.

sequoiae
n 27
number of 22-32
flagellomeres 23
body length 2.5-4.1
(mm) 24
FWL (mm) 2.1-3.6
2.4
MS/EH 0.41-0..60
0.50
TW/EH 0.58-0. 85
0.68
IOD/EH 1.00-1. 35
1.18
OTW/OED 0.67-0. 92
0. Tt
OED/OOD 0. 71-1. 00
0.87
Pet. L/(2-3)L L27-1..72
Leth
ICu,/Ro 0.96-1.27
1.00
OSL/FWL 0. 33-0. 80
0. 63

canadensis

18

21-28
broken

1071.35

aphenges

0.49-0.55
0.52

0. 60-0. 79
0.67

1.20-1.43
1.24

0.64-0. 82
0.67

0.81-1.00
0.81

1.22-1. 30
1.22

0.91-1.00
0.95

0.32-0.46
0.43

Color: Body color highly variable from more or less wholly testaceous to
piceous, but commonly dark ferrugineous, the head very slightly lighter; typi-
cally, the apical 1/3 of the petiole and basal 1/2 of tergite (2-3) are distinctly
lighter than the remainder of the gaster and thorax; legs usually lighter than
body, especially in darker specimens, hind coxae and legs usually a shade

46 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

darker than the fore and mid coxae and legs, the tibiae usually distinctly lighter
banded basally; antennae usually testaceous basally, becoming tan to brown api-
cally; maxillary and labial palpi light testaceous; eyes grey, with no trace of
iridescence in white light; tegulae testaceous, usually lighter than thorax; fore
wings weakly infumated, at most lightly banded (Fig. 27) to subhyaline; veins
and stigma tan to brown, the basal 1/3 of stigma lightened. (The wings have
also been figured by Marsh, 1963 and 1965, under S. brunneri. )

Males: Essentially agree with the females in color and sculpture, parallel-
ing the variation noted, though the sculpture, particularly of the frons, vertex
and mesopleural disc tends to often be greatly reduced in intensity; however,
sculpture on base of tergite (2-3) when present may cover up to the basal 1/2; .
also the areola of the male propodeum is often greatly reduced compared to
that of the female, and the basal carina thus very long, much longer than its
forks; male tarsi (Fig. 16) are proportionately distinctly shorter than those of
the female, and hind tarsomere 3 is distinctly shorter than tarsomere 5; size
ranges from 1.5-3.7 mm; Marsh (1965) has figured the genitalia.

Remarks: Spathius sequoiae was described by Ashmead (1889) from two
females, one of which bears the label ‘‘Spathius sequoiae Ashm.’’ in Ash-
mead’s handwriting. I have designated this specimen as lectotype and the
other becomes a paralectotype; both bear three additional labels ‘‘Alameda Co.
Cal.’’, ‘‘Parasite on Col. larva on redwood’’, and the red ‘‘USNM Type No.
puco .

In addition to the types of each of the original three species, I have seen
several short series of reared specimens from various species of coniferous
trees in western North America, and measured about 40 selected individuals of
both sexes. I have been unable to find any consistent differences by which
Specimens may be separated, and it appears that the chief way in which they
were distinguished in the past was by sorting according to host. Spathius cali-
fornicus, based on a single male, has always been an uncertain species, and _
was never formally described beyond the characters given by Ashmead (1893)
in his ‘‘Table of Species’’. It was distinguished from 8S. sequoiae in opposite
parts of his table on whether the wings appeared hyaline (californicus) or
banded (sequoiae), an inadequate character. Both are from adjacent localities
in California (Santa Cruz Mts. and Alameda Co.) and of opposite sex, a situa-
tion immediately suspect. Viereck (1912) described Spathius brunneri from two
females reared from Pissodes in Pseudotsuga from Columbia Falls, Montana
(Hopkins No. 8574). He failed to note its similarity to sequoiae, instead think-
ing it closest to S. unifasciatus Ashmead (=trifasciatus Riley), comparing the
two in his description. Actually, they are quite different, having only super-
ficial similarity in size and body coloration, particularly in the lightened area
at the apex of the petiole and base of tergite (2-3). The very stout petiole of
the type is atypical, that of the paratype being the more usual case.

Diagnosis: A highly variable species, S. sequoiae may be characterized
by the weakly infumated fore wings (Fig. 28), well developed lateral areas of
the prothorax, delicately transversely striate vertex, weak to almost obscure
sculpture on about basal 3/8 of tergite (2-3), and the form of the sternaulus and
mesopleural disc. It is most closely related to S. canadensis, from which it
may usually be distinguished by the longer ovipositor (0. 60-0. 80x FWL versus
0.45-0. 65x FWL in canadensis) and presence of more spines on the fore tibia
(15-33 in irregular double and sometimes triple rows versus 11-23 in irregular
single to double rows in canadensis). Qualitatively, canadensis, typically a
smaller and more slender species, tends to have a more scaly reticulate meso-
pleural disc than sequoiae, and also may show additional weak sculpture follow-

Matthews: Nearctic Spathius 47

ing the basal 3/8 sculpture on tergite (2-3). Males of the two differ additionally
in relative lengths of tarsi to tibiae, sequoiae males having relatively shorter
tarsi (Fig. 16) (hind tarsus/tibia ratio of 0.66-0.88, versus 0. 84-0. 96 in
canadensis males) but small specimens may be virtually inseparable.

Indeed, until additional reared series and biological information are avail-
able, I regard this separation as tentative, but prefer to retain the two as full
species rather than subsets of a polytypic species. Both sequoiae and cana-
densis are parasites of scolytids occurring in coniferous trees, canadensis
being commonly reared from spruce in eastern North America and sequoiae
from a wide variety of coniferous trees (including spruce) in western North
America. A series from Picea glauca from Seebe, Alberta (CNC) possess
some apparently intermediate characters. Another from redwood in Mill
Valley, California (CAS) includes two females with very short ovipositors
which would key to aphenges but may be distinguished by the greater petiole to
tergite (2-3) ratio (1.27-1.72 versus 1.22-1.30 in aphenges) and the absence
of sculpture beyond the basal 3/8 of tergite (2-3).

Distribution: Widespread throughout western North America, north to
Alaska. I have seen material from Alaska, California, Colorado, Idaho,
Montana, Oregon, New Mexico, Utah, Washington, British Columbia and
Alberta.

Hosts: No hosts were previously recorded for Spathius californicus.
Spathius sequoiae is recorded as a parasite of Phloeosinus sequoiae Hopkins
and P. punctatus LeConte, both scolytids (Muesebeck et al., 1951, DeLeon,
1952, and Bushing, 1965). I have confirmed these records and the specimens
on which they are based bear Hopkins Nos. 15815a and 18161-b. Spathius
brunneri is recorded from Pissodes sp. (Curculionidae) and the scolytids
Pseudohylesinus nebulosus (LeConte) and Scolytus ventralis LeConte (Viereck,
1912, Bedard, 1938, Muesebeck et al., 1951, Bushing and Bright, 1965,
Bushing, 1965, Muesebeck, 1967, and Asraf and Berryman, 1969). Ihave also
confirmed these records.

According to the material before me, S. sequoiae has been reared from
scolytids in the following trees: Pseudotsuga, Tsuga, Chamaecyparis, Larix,
Sequoia, Thuja, Abies and Picea. The majority of the records are from
Pseudotsuga; associated insects are listed under Hopkins Nos. 8574, 10764b,
20225, 31532-F, 31713-D, 35312-E, 353560 and 39696b. Additional hosts
not recorded above include the scolytids Dendroctonus pseudotsugae Hopkins
and Scolytus tsugae Swaine. Host records from Tsuga include Scolytus sp.,
recorded under Hopkins Nos. 1849n2, 2174e5, 2174g2a and 19858. The single
record from Chamaecyparis (Hopkins No. 33316c) lists Phloeosinus sequoiae
and P. punctatus as associated insects. The records from Sequoia and Thuja
are listed above as the hosts of Spathius sequoiae. Hopkins No. 34214B-7 re-
cords Dendroctonus obesus as the host in Picea engelmanni. According to
D. F. J. Hilton (pers. comm.) the series he reared from Picea glauca in
Alberta was associated with the following scolytids: Carphoborus intermedi-
us, Dryocoetes affaber, Pityophthorus murrayanae, Polygraphus rufipennis,
Scolytus piceae and Trypodendron lineatum. No hosts are recorded from Larix.

10. Spathius canadensis Ashmead

Spathius canadensis Ashmead, 1891, Can. Ent. 23:2. [Type locality: Canada:
Ottawa (holotype female in USNM, Type No. 52649)].

Spathius claripennis Ashmead, 1893, Can. Ent. 25: 70, 72. [Type locality:
West Virginia: Morgantown (lectotype male in USNM, Type No. 2148)].
New synonymy.

48 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

Spathius clavipennis [sic] Ashmead, in Hopkins, 1892, Insect Life 4: 257.

Spathius tomici Ashmead, 1893, Can. Ent. 25: 71. [Type locality: West
Virginia: Morgantown (holotype female in USNM, Type No. 52648)].
New synonymy.

Females: Body length, 2.0-3.8 mm; FWL, 2.0-2.8 mm; ovipositor sheath
length, 0.95-1.7 mm. Other measurements and ratios in Table 3.

Head: Flagellomeres 21-28; face transversely strigose-granular; frons
varying from almost perfectly smooth to finely transversely striate; vertex
smooth and shining to delicately transversely strigose; temples typically with
faint irregular sculpture posteriorly adjacent to occipital carina; ocellar tri-
angle slightly raised, smooth to faintly strigose-granular.

Thorax: Lateral areas of prothorax moderately well developed extending
to precoxal cavities and with several irregular oblique cross carinae; notauli
distinct, shallowly crenulate anteriorly, meeting in a slightly depressed weak-
ly rugose area from which two weak carinae usually extend to the posterior
margin of the mesoscutum; mesonotum abruptly declivous, the lobes evenly
scaly-reticulate to granular; scutellar furrow variable, with as many as 8
cross carinae but often with a single well developed medial carina flanked by
weaker ones; scutellar disc scaly-reticulate to weakly granular; mesopleural
disc variable, either nearly smooth above sternaulus or weakly scaly-reticulate
to delicately strigose granular; sternaulus highly variable, from strongly im-
pressed and crossed by several fine closely spaced carinae, the more posteri-
or elements apparently continuing and curving posteriorly to mesocoxal cavities,
to almost obscured, scaly-reticulate, the cross carinae indistinct or irregular
and none distinctly curving posteriorly to mesocoxal cavities; prepectal carina
usually complete but often weak or nearly obscured immediately above sternau-
lus; prepectal area more or less smooth with sparse setae and sometimes with
weak irregular surface rugosity; propodeal carinae typically distinct and well
defined; dorsal areas granular to weakly rugose; areola usually trapezoidal,
slightly longer than broad (sometimes very elongate), enclosing weak irregular
sculpture; basal carina usually not longer than its forks; fore tibia with 11-24
spines arranged in irregular single to double rows and a row of 5-7 spines at
apex; outer apical lobe of hind tibia with a row of 3-5 spines; hind coxae weakly
obliquely striate dorsally; tibiae and tarsi usually very slender; subdiscoideus
not interstitial.

Abdomen: Petiole irregularly longitudinally strigose-granular and arched
at base; epipleurite of petiole variable, sometimes more than half of the peti-
ole length; tergite (2-3) usually with distinct area of granular to scaly-reticulate
sculpture on about basal 3/8 to 4/9, with or without an additional very weak
scaly-reticulate area following a narrow unsculptured line; remaining tergites
typically smooth, but sometimes with sparse punctulation anterior to the trans-
verse row of setae; epipleurites usually very faintly punctulate; lateral margin
of tergite (2-3) usually thickened weakly at base; ovipositor slightly curved,
usually 0. 45-0. 62x FWL.

‘Color: Body color usually uniformly dark shining testaceous, the gaster
beyond the basal half of tergite (2-3) usually distinctly darker; legs concolorous
with body or slightly lighter, hind coxae and legs sometimes a shade darker
than fore and middle coxae and legs, the tibiae indistinctly lighter banded at
bases; eyes grey to faintly iridescent green in white light; antennae testaceous,
often lighter than head at base, becoming light brown apically; ovipositor sheaths
usually brown, sometimes contrasting strongly with the lighter gaster; fore
wing infumation variable, usually weakly infumated to sub-hyaline, but one

Matthews: Nearctic Spathius 49

series from Wood Co., Wisconsin (RDS) has several individuals with well de-
fined fasciations; veins and stigma tan or stramineous, the basal 1/3 of stigma
usually lighter.

Males: Essentially agree with the females in color and sculpture, parallel-
ing the variation noted; size smaller, 1.6-2.5 mm, and number of flagello-
meres varies from 19 to 25; body color often much darker than for females, to
nearly piceous, except lighter base of gaster and legs; wings usually hyaline to
sub-hyaline; sculpture on base of tergite (2-3), when present, may cover as
much as 1/2 of tergite; tarsi not especially shortened, hind tarsus to tibia
ratio of 0. 84-0. 95.

Remarks: Since 1936, Spathius canadensis Ashmead has become firmly
entrenched in the literature and almost automatically applied to any Spathius
reared from elm. However, this association is incorrect (see ‘‘Remarks’’ for
S. benefactor). Detailed study of Ashmead’s type reveals it to agree most
closely with that of S. tomici and with females associated with the type male of ,
S. claripennis. Ashmead’s concept of canadensis, based upon a single female
from an unknown host, was insufficiently stringent, for specimens bearing his
canadensis determination labels belong to at least three different species. It
is easy to appreciate how confusion could arise, for the type belongs to a dif-
ficult complex of species and bears superficial resemblance to several other
species. These misplaced canadensis specimens were probably identified on
the basis of their overall ‘‘gestalt’’: chiefly their relatively small size, short
ovipositor and sculpture of tergite (2-3).

The female of S. claripennis* was previously unknown; I base my associa-
tions on three short series of both sexes from spruce: one from Webster, New
Hampshire (Hopkins No. 3726c), another from Cherry Mt., New Hampshire
(Hopkins No. 17614c) and the third from Rensselaerville, New York (MCZ).
The males from these series closely match Ashmead’s type of claripennis,
which is an unusually small individual, and the females essentially agree with
the type of S. tomici, described in the same paper. The type of tomici appears
superficially different due to the strongly arched petiole with its epipleurite
extending over half of the petiole length, but these characters are highly vari-
able. The type of S. canadensis falls readily within the range of variation
exhibited by claripennis and tomici. At first glance it might appear different
due to the presence of weak punctulation which it exhibits anterior to the setal
rows on tergites 4-7. However, by itself punctulation is an unreliable char-
acter, for commonly it is variously expressed within series from several
Spathius species. Furthermore, even when present it may not be readily ap-
parent unless the gastral tergites are somewhat distended to expose the area
anterior to the transverse row of setae on each.

Diagnosis: Spathius canadensis may be distinguished from its closest rela-
tive, S. sequoiae, by its generally smaller, more slender body and smaller

oops claripennis was apparently based on two males, according to numbers on
Hopkins’ original manuscript for his 1892 paper. The specimen bearing the number
2463, according to Hopkins’ paper, emerged on April 20, 1892, agreeing with
Ashmead’s statement in his description. A second individual with the same rearing
information but bearing number 2463a is also listed in the original manuscript with no
date of emergence given and with the name Spathius claripennis written in the margin.
Accordingly, I designate as lectotype the specimen bearing number 2463 and the other
specimen (2463a) is thus a paralectotype.

50 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

OSL/FWL ratio (0.45-0. 65 versus 0. 60-0. 80 in sequoiae). Additional charac-
ters are given in the key; see also ‘‘Diagnosis’’ for S. sequoiae.

Distribution: Probably widespread throughout northern North America,
south to the higher elevations of North Carolina. I have seen specimens from
the following states and provinces: Alberta, British Columbia, Maine, Mani-
toba, Michigan, New Hampshire, New York, North Carolina, Ontario, South
Dakota, Washington, West Virginia and Wisconsin.

Hosts: Spathius canadensis, as defined here, is apparently parasitic upon
Scolytidae attacking the smaller branches and crown of spruce, fir, larch and
pine. The following are regarded as valid host records: Polygraphus rufipen-
nis, Dryocoetes autographus, Ips pini, Pityophthorus balsameus, Orthotomicus
caelatus, Scolytus piceae and Cryphalus sp.

Unfortunately, due to the confusion surrounding the identity of S. canadensis
in the past, there have been falsely recorded, under this name, at least 19 hosts
representing four coleopterous families from a diverse range of trees. (Fora
fairly complete summary, see Bushing, 1965.) In the following paragraphs, the
interested reader will find a discussion of these.

The first hosts for S. canadensis were reported by Hopkins (1892) and in-
cluded three bark beetles, Phloeosinus graniger (= dentatus), Tomicus sp. and
Dryocoetes americanus (= autographus autographus), and the weevil Magdalis olyra, from
red cedar, white pine, fir and hickory, respectively; the canadensis speci-
mens correlated with these bear, according to Hopkins’ original manuscript,
the numbers 2932 (and 2674, 3066), 2589, 2442 and 5180, respectively. Of
these, I have been able to locate only 2442, which is S. brachyurus; although
unconfirmed, those from Phloeosinus would almost certainly be S. impus.

The following year, Hopkins (1893b) added the buprestid Agrilus otiosus
from hickory to the list, and furnished specific identification for the Tomicus
(= Ips grandicollis), but made no mention of either Dryocoetes or Magdalis.
Unable to associate names with specimens, I suggest they be doubted, espe-
cially Agrilus which was probably a host of S. simillimus.

In 1899, Hopkins added the weevil Copturus (= Cylindz Cylindrocopturus) longulus
from pine, another record that Iam unable to confirm.

Scolytus rugulosus in apple was recorded as a host of canadensis by Schedl
(1932) based on a determination by A. D. Cushman. I suspect that these speci-
mens will probably be found to be S. benefactor.

MacAloney (1930) records Pissodes strobi (Curculionidae) which seems
possible, but is again unconfirmed.

In 1936, Kaston and Becker added three new hosts from elm: Hylurgopinus
rufipes, Magdalis barbita and M. armicollis. I have seen their specimens and
many subsequent series reared from elm, and all should be henceforth recorded
as hosts for the new species, Spathius benefactor. Pechuman (1937) added
Scoytus multistriatus and Saperda tridentata from elm, and Hoffman (1938) re-
corded Leperisinus aculeatus from ash; these should also be changed to hosts
of S. benefactor.

~ Hetrick (1940) listed Dendroctonus frontalis from pine; this should be
transferred to Spathius pallidus. Beal and Massey (1945) report canadensis
from Scolytus quadrispinosus in hickory, probably erroneous, although I have
been unable to locate their specimens.

In summary, of all the hosts previously listed for Spathius canadensis,
none can be confirmed for this species as it is presently defined.

The only host record previously recorded for Spathius claripennis was
Ashmead’s original record from the scolytid Polygraphus Ss rufipennis under
bark of Abies nigra; subsequent references to this host are listed by Bushing

Matthews: Nearctic Spathius D1

(1965) under the name P. poligraphus L.

The following scolytid hosts are recorded for Spathius tomici: Dryocoetes
autographus from spruce (Hopkins, 1893a); Pityophthorus balsameus from
balsam fir (Felt, 1906); Orthotomicus caelatus and Scolytus piceae from larch
(Blackman and Stage, 1918); Phloeosinus canadensis and P. dentatus from ce-
dar (Schedl, 1932, and Beal and Massey, 1945). Subsequent references to each
are listed by Bushing (1965). Of these, I have seen only specimens of Schedl
and Beal and Massey; both records should be transferred to Spathius impus.
The others, except possibly Dendroctonus simplex, seem reasonable insofar
as the species is currently defined.

Previously unrecorded hosts include the scolytids Cryphalus sp. (Hopkins
No. 2871) and Ips pini (Say). Ihave reared S. canadensis in New York
from white spruce in association with Polygraphus rufipennis (Kirby), Dryocoe-
tes autographus (Ratz.) and Crypturgus atomus LeConte (det. D. E. Bright), all
Scolytidae.

11. Spathius aphenges, new species
(Fig. 26)

Holotype. - female, California: El Dorado Co., xi-1963 [no collector label],
ex, Pinus lambertiana with Orthotomicus latidens [USNM Type No.

70502].

Description of type female: Body length, 2.5 mm; FWL, 2.3 mm; oviposi-
tor sheath length, 1.0 mm. Other measurements and ratios in Table 3.

Head: Flagellomeres 23; face shining almost smooth except for weak trans-
verse rugosity; frons nearly smooth and shining with very faint transverse stri-
ation; vertex smooth and polished; temples smooth; ocellar triangle weakly
raised, smooth.

Thorax: Lateral areas of prothorax well defined, relatively broad and
crossed by several oblique carinae; remainder of propleura finely horizontally
strigose to smooth; notauli shallowly crenulate anteriorly, meeting ina small
slightly depressed area of weak rugosity; mesonotum moderately declivous, the
lobes evenly granular; scutellar furrow lacking sharp well defined cross cari-
nae, but with 3-5 irregular variously developed carinae; scutellum slightly
raised, the disc evenly granular; mesopleural disc smooth and shining, becom-
ing abruptly rugose on dorsal anterior part above prepectal area; sternaulus
deeply impressed with about 7 poorly defined cross carinae, the posterior mar-
gin clearly delimited and followed by a smooth space to mesocoxal cavities;
mesosternum smooth; prepectal carina complete; prepectal area covered by
sparse setae, otherwise apparently smooth with faint basal irregularities;
propodeal carinae sharp and well defined; dorsal areas granular to weakly ru-
gose; basal carina about as long as its forks; costulae inserted at about right
angles to basal forks; areola approximately diamond-shaped, about 1. 6x as long
as basal carina and enclosing weak rugosity; fore tibia with 14 spines arranged
in a single row with a row of 6 spines at apex; outer apical lobe of hind tibia with
a row of 5 spines; hind coxae with weak oblique striae dorsally; subdiscoideous
not interstitial.

Abdomen: Petiole arched basally, dorsally coarsely granular to rugose,
becoming longitudinally strigose apically; tergite (2-3) prominently granular
on basal 4/9, followed by a narrow smooth area and then 2/9 scaly-reticulate
area of much less intensity, the apical 1/3 smooth; lateral margin of tergite

52 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

(2-3) weakly thickened at base; ovipositor much shorter than abdomen, about
0.43x FWL.

Color: Head shining light brown weakly suffused with dirty yellow in malar
area and lower clypeus; prothorax dirty testaceous; remainder of thorax and
propodeum light brown, the propodeum slightly lighter; petiole and tergite
(2-3) dirty testaceous, remainder of gaster slightly darker, almost a brown;
ovipositor sheaths testaceous at base, becoming brown apically; antennae
testaceous basally, decidedly lighter than head, becoming light tan apically;
maxillary and labial palpi, tegulae, coxae and legs wholly light testaceous,
distinctly lighter than body; eyes gray, not at all iridescent in white light;
wings weakly infumated as in Figure 26; veins testaceous to tan, stigma light
brown, the basal 1/3 lighter stramineous.

Variation: Paratype females. Body color varies from about wholly
testaceous to nearly ferrugineous; appendages (except ovipositor sheaths) of
lighter individuals may not particularly contrast with body as in the type;
eyes never iridescent; number of fore tibial spines varies from 11-17; sculp-
ture of frons may be totally obscured in some individuals and it is never
stronger than in type; vertex and temples nearly always smooth, though occa-
sionally faint irregularity occurs; sculpture on basal part of tergite (2-3) always
present, though sometimes reduced in intensity, but sculpture on middle part of
tergite is often obscured or only faintly scaly-reticulate; sternaulus develop-
ment somewhat variable, sometimes barely developed with no obvious cross
carinae, the whole of the mesopleuron (except dorsal part) appearing smooth;
fore wings of some individuals subhyaline; variation in measurements and
ratios given in Table 3.

Paratype males: The two paratype males agree essentially with the females.

Paratypes: 19, 10’, same data as holotype. [USNM]. U.S. A., California:
Strawberry, 19, 22 July 1962, Flight trap, Hopkins No. 37714GGG (W. D.
Bedard) [USNM]. Oregon: Ashland, 499, lo’, 4 June 1914, reared Pinus lam-
bertiana, Hopkins No. 10884d (P. D. Sargent) [USNM].

Remarks: The small size and very short ovipositor (less than 0.5x FWL)
immediately distinguish this species and Spathius parvulus from all the other
Nearctic species. The combination of the smooth, shining vertex, non-
iridescent eyes and only weakly infumated fore wings (compare Figs. 26 and
30) will usually readily distinguish aphenges from parvulus. In addition,
aphenges has a relatively shorter petiole (ratio of petiole length to tergite
(2-3) length, 1.22-1.30 in aphenges versus 1. 33-1.45 in parvulus) and a
broader temple (TW/EH, 0. 60-0. 79 in aphenges versus 0. 50-0. 63 in parvulus).
Occasional individuals of S. sequoiae having ovipositors less than 0.5x FWL
may be distinguished by the characters discussed in the ‘‘Diagnosis’’ section
under that species.

Distribution: Probably widespread in western North America. In addition
to the paratype localities, I have seen specimens from Idaho, Washington,
Alberta and British Columbia.

Hosts: The type series was reared from the scolytid Ips latidens LeConte
in sugar pine, Pinus lambertiana. The paratype series from Oregon was also
reared from P. lambertiana, according to Hopkins No. 10884d, and were ‘‘ex.
2 cones of 1913 crop killed by cone beetle.’’ Godwin and Odell’s (1965) refer-
ence to Spathius sp. emerging from cones infested with this beetle (Conoph-
thorus coniperda (Schwarz)) no doubt refer to this species, although I have not
seen their specimens. A specimen from Alberta is labeled as being from the
scolytid Phthorophloeus puberulus (LeConte), which according to Chamberlin
(1939) attacks the tops and limbs of Douglas fir, Pseudotsuga taxifolia. I have

Matthews: Nearctic Spathius D3

also seen a series of Spathius aphenges reared from Scolytus abietis Blackman
in grand fir, Abies grandis, from Idaho. In addition, single females were
reared from Pinus contorta in British Columbia and P. ponderosa in Oregon
(Hopkins No. 10074), the latter in association with several borers and the

scolytid Pityophthorus sp.

12. Spathius pallidus Ashmead
(Fig. 28)

Spathius pallidus Ashmead, 1893, Can. Ent. 25: 72, 74. [Type locality:
West Virginia: Morgantown. (holotype female in USNM, Type No.
2152)].

Spathius pollidus [sic] Ashmead, in Hopkins 1893a and 1893b, Bull. W. Va.
agric. Exp. Stn. 31: 145 and 32: 346.

Females: Body length, 2.2-3.8 mm; FWL, 1.9-3.2 mm; ovipositor
sheath length, 1.2-2.1 mm. Other measurements and ratios in Table 4.

Head: Flagellomeres 26-33; face coarsely transversely strigose-granular,
ending abruptly at malar space; eyes very slightly emarginate adjacent to an-
tennal insertions; frons transversely striate; vertex uniformly transversely
strigose, becoming less strong or scaly-reticulate to nearly smooth at eyes;
temples smooth along eyes, becoming moderately strigose posteriorly, more
or less parallel to occipital carina; malar space rarely greater than 1/2 EH;
ocellar triangle broad, weakly raised and delicately granular to smooth.

Thorax: Lateral areas of prothorax moderately developed, losing distinct-
ness posteriorly in the irregular strigose-granular sculpture of remainder of
prothorax; notauli distinctly impressed and crenulate anteriorly, merging toa
broad, slightly depressed usually moderately rugose area; mesonotum moder-
ately declivous, robust, greatest width at least equal to greatest length, usual-
ly slightly broader than long, the lobes evenly granular; scutellar furrow
crossed by 3-10 more or less evenly spaced straight carinae; scutellum broad,
weakly projecting, the disc evenly granular right to the truncate posterior mar-
gin, there never being any rugosity posteriorly; mesopleural disc rarely partly
smooth, but typically strigose-granular above sternaulus, this becoming much
more pronounced dorsally below wings; sternaulus broad, usually well im-
pressed, crossed by several closely spaced variously developed carinae, many
of which may appear to extend onto mesopleural disc; the space posterior to
the sternaulus usually with 2-6 curved striae extending from the posterior end
of the sternaulus to the mesocoxal cavity; prepectal carina weak immediately
above the sternaulus; prepectal area nearly smooth, with weak surface irregu-
larity; propodeal carinae all more or less distinct, often with numerous short
side branches; dorsal areas granular to weakly rugose; basal carina usually
shorter than its forks; the segments of the basal forks anterior to the costulae
insertions about 2-3x the posterior segments, the areola thus an elongate trape-
zoid enclosing an area of irregular weak reticulation; area petiolaris rectangu-
lar, about 1/3 as long as areola and lacking distinct sculpture centrally; fore
tibia with 18-25 spines arranged in an irregular double row with a row of 5-8
spines at apex; outer apical lobe of hind tibia with a row of 3-5 spines situated
somewhat anteriorly; claws often with prominent pulvilli; subdiscoideus not
interstitial; first intercubitus typically slightly longer than second abscissa of
radius.

Abdomen: Petiole arched at base, dorsally coarsely granular to longitu-

54 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

dinally strigose; tergite (2-3) with a distinct area of granular sculpture on about
basal 2/5 to 5/11, followed almost immediately by another area of less well
developed granular or scaly-reticulate sculpture extending nearly to transverse
row of setae, the apical margin beyond the setal row smooth; remaining tergites
vary, frequently apparently smooth, but often with weak transverse bands of
faint scaly-reticulation or sparse punctulation anterior to the setal rows on at
least the 4th tergite; ovipositor from 0. 60-0. 72x FWL.

Color: Body color typically wholly tawny yellow to light brown, but varying
from light honey yellow to light ferrugineous, the gaster beyond the basal 2/3
of tergite (2-3) sometimes slightly darker; coxae and trochanters concolorous,
usually a shade lighter than the body, from stramineous to light testaceous;
femora and tibiae concolorous with body, the basal 1/3 to 1/4 of tibiae lighter
banded; tarsi testaceous, the ultimate tarsomere and claws usually slightly
darker than others, typically dull ferrugineous; antennae testaceous to light
brown, darkening gradually toward apex; ovipositor sheaths testaceous at base,
becoming dark brown on about apical 1/3; eyes light gray, sometimes with a
weak iridescent hue in white light; fore wing darkly infumated and distinctly
banded (Fig. 28); veins light to dark brown; stigma brown, the basal 1/3 lighter,
dirty yellow to stramineous.

Males: Agree with females in most respects and parallel the variation noted,
except that tergite (2-3) has a larger area of well developed granular sculpture
covering about basal 1/2 followed by another area of less strong sculpture as in
female, and the areola is often reduced in size, sometimes to the form of a
small diamond, the basal carina thus longer than its forks; also, small indivi-
duals sometimes have marked reduction in sculpture, particularly that of ver-
tex which may appear nearly smooth.

Remarks: The holotype, USNM Type No. 2152, bears the printed label
‘¢3768 Hopk. W.Va.’’ which is not in Hopkins’ original manuscript for his 1892
paper. Nor could I find it in the Hopkins card file. It also bears Ashmead’s
handwritten label ‘‘Spathius pallidus Ashm.’’ The type catalog says two
specimens were included, but I have only located the one, and consider it the
holotype. It is lacking the antennae and both fore wings.

Diagnosis: Spathius pallidus is best distinguished by the combination of the
typically wholly tawny yellow body with the gaster at most slightly darker be-
yond the base of tergite (2-3), the robust mesonotum and strigose sculpture on
the mesopleural disc. It most resembles S. impus and S. parvulus, sharing
the narrow temples and narrow striated vertex. It may be distinguished from
impus by its greater OTW/OED ratio (0. 80-0. 93 versus 0. 60-0. 83 in impus)
and the presence of additional sculpture on the middle portion of tergite (2-3)
which is absent in impus. S. parvulus differs in having a shorter ovipositor
(typically less than half of FWL compared to 0. 60-0. 70 in pallidus) and a
smooth mesopleural disc; also the form of the vertex sculpture differs (see
Fig. 25).

Distribution: Southeastern United States north to Massachusetts. I have
seen material from the District of Columbia, Georgia, Louisiana, Maryland,
Massachusetts, Mississippi, North Carolina, South Carolina, Texas and
Virginia.

Hosts: Four hosts are recorded, the scolytids Ips grandicollis (Eichhoff)

(= Tomicus cacographus LeConte) and Dendroctonus frontalis Zimmermann,
the buprestid Chrysobothris femorata (Oliv. ) and the cerambycid Callidium
variabilis L. A resumé of the literature references to these hosts is given in
Bushing (1965). Based on the material I have seen, the latter two hosts are
probably incorrect, being most likely based on misidentifications of Spathius

Matthews: Nearctic Spathius 9)

simillimus or S. floridanus.

Much of the material I have seen is reared under various Hopkins numbers,
all from various species of Pinus. The most frequently cited host is D.
frontalis (Hopkins Nos. 10099v, 46768, 46847, and 17843dib). New host records
include the scolytid Ips avulsus (Eichhoff) (Hopkins No. 3350j) and the weevils,
Pissodes nemorensis Germ. (Hopkins No. 6885h) and P. approximatus Hopkins
(Hopkins No. 3810d). Hetricks’ (1940) reference to Spathius canadensis as
parasitic on Dendroctonus frontalis ( and subsequent references to this record
in Bushing, 1965) should be changed to this species, as I have seen his speci-
mens.

13. Spathius parvulus, new species
(Figs. 25, 30)

Holotype.- female, New Mexico: Vermego, Bred Sept. 11, ’08, Pinus
ponderosa, Hopk. U. S. 2410a (A. D. Hopkins) [USNM Type No. 70503].

Description of type female: Body length, 2.0 mm; FWL, 1.9 mm; oviposi-
tor sheath length, 0.75 mm. Other measurements and ratios in Table 4.

Head: Flagellomeres 21; face delicately strigose-granular; frons delicately
transversely striate; vertex delicately strigose, the striae curving medially and
anteriorly toward ocellar triangle (Fig. 25); temples apparently smooth with
extremely faint irregularity posteriorly near occiput; ocellar triangle weakly
raised with delicate granular sculpture which continues weakly onto vertex.

Thorax: Lateral areas of prothorax shallow, but distinct with several
oblique cross carinae on smooth background; remainder of propleura finely
granular; notauli shallowly crenulate anteriorly, losing definition posteriorly
except for faint irregular rugosity in area of convergence; mesonotum robust,
strongly declivous, the lobes evenly granular; scutellar furrow with 4 evenly
spaced carinae; scutellum weakly convex, the disc uniformly granular; meso-
pleural disc apparently smooth or faintly scaly-reticulate, with delicate, pre-
dominantly horizontal striae dorsally; sternaulus a broad, shallow depression
with obscure traces of a few cross carinae; prepectal carina complete; prepectal
area Smooth, sparsely hairy with some irregular rugosity ventrally; propodeal
carinae all distinct and well developed; dorsal areas wholly granular; basal
carina slightly shorter than its forks; costulae approximately perpendicular to
basal forks; areola and area petiolaris distinctly separated, the areola about
1.6x as long as area petiolaris and granular; area petiolaris rectangular with
one carina across basal third; fore tibia with 18 spines arranged in an irregular
single row with a row of 6 spines at apex; outer apical lobe of hind tibia with 3
spines; tarsomere 5 distinctly longer than tarsomere 3 on all legs; subdiscoide-
us not interstitital; recurrent vein distinctly longer than first abscissa of radius.

Abdomen: Petiole about as long as middle tibia and strongly arched at base;
petiole sculpture granular dorsally at base becoming irregularly longitudinally
strigose at apex; tergite (2-3) with prominent granular to scaly-reticulate sculp-
ture on about basal 2/3, the remaining 1/3 smooth; sculptured portion of two
distinct intensities, basal 3/5 much stronger, abruptly becoming much less
intense for remainder; lateral margin of tergite (2-3) weakly thickened at base;
remaining tergites smooth; ovipositor much shorter than abdomen.

Color: Body color wholly testaceous except for gaster beyond basal 2/3 of
tergite (2-3) which is dull ferrugineous; ovipositor sheaths brown; legs uni-
colorous, testaceous, the tibiae not distinctly banded at base, the claws and

06 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

ultimate tarsomeres Slightly darker; antennae testaceous, gradually becoming
light brown apically; palpi testaceous; eyes with a distinct bluish-green irides-
cent hue in white light; wings lightly infumated as in Figure 30.

Variation: Paratype females. The three paratype females show no signifi-
cant variation, although the intensity of the vertex striation shows some ten-
dency for reduction (but still discernable); number of fore tibial spines varies
from 16-18. Variation in measurements and ratios given in Table 4.

Paratype males: Similar to females in all respects except that sculpture
on tergite (2-3) is much stronger and of approximately uniform intensity through-
out, covering about basal 3/4 of the tergite; however, vertex striation tends to
be considerably reduced in intensity to almost smooth; lateral margin of most
tergites thickened; junction of costella and radiella of hind wing slightly thick-
ened.

Paratypes: 399, 3o’o’, same data as holotype.

Remarks: This species is apparently related to Spathius impus and shares
with it the relatively narrow vertex and temples. It differs from it chiefly by
the shorter ovipositor (typically less than 0.5x FWL versus 0.5-1.02x FWL in.
impus) the more extensively sculptured tergite (2-3), and the form of the ver-
tex sculpture (Fig. 25), which also distinguishes it from the other Nearctic
species. About two dozen specimens from several localities in eastern United
States are tentatively included here, but may later prove to be a distinct spe-
cies. These are similar to the type series, but have the face more coarsely
granular, the vertex simply transversely strigose, and are generally darker
colored.

Distribution: The type series is from Vermego, New Mexico. I was unable
to find this locality in Hammond’s New World Atlas and think it is likely a mis-
spelling of Vermejo, north of Maxwell in Colfax County. I have also seen
specimens from South Dakota and Colorado. Specimens from eastern North
America tentatively placed here are from Maine, Maryland, Massachusetts,
New Hampshire, New York and West Virginia.

Hosts: The type series was recorded under Hopkins No. 2410a which lists
Tomicus [=Ips] sp. in ponderosa pine as the apparent host. Three specimens
from Winchester, Massachusetts (N. E. For.Ins. Lab No. 12472y2) are labeled
‘Cex, Pissodes strobi’’ (white pine weevil).

14. Spathius impus, new species
(Fig. 18)

Holotype.- female, Connecticut: East River, reared 31 May 1916, Hopkins No.
10762a, Juniperus virginiana (C. R. Ely) [USNM Type No. 70504].

Description of type female: Body length, 2.9 mm; FWL, 2.5 mm; oviposi-
tor sheath length, 1.7mm. Other measurements and ratios in Table 4.

Head: Flagellomeres 24; face transversely strigose-granular; frons weakly
transversely striate; vertex transversely strigose; temples smooth adjacent to
eyes, becoming weakly wrinkled strigose posteriorly; ocellar triangle weakly
granular, very Slightly raised.

Thorax: Lateral areas of prothorax weakly developed, shallow, smooth
with a few indistinct delicate oblique cross carinae, becoming obliterated in the
finely granular-strigose sculpture of remainder of prothorax; notauli distinct,
shallowly crenulate anteriorly, meeting in a slightly depressed area with weak
surface irregularity dominated by two elongate subparallel longitudinal carinae

Matthews: Nearctic Spathius 57

which extend to the posterior margin of mesonotum; mesonotum rather abruptly
declivous, the lobes evenly granular; scutellar furrow narrowest at mid-point,
with apparently 5 cross carinae; scutellar disc uniformly delicately granular
or scaly-reticulate; mesopleural disc delicately granular or scaly-reticulate
above sternaulus, becoming irregularly longitudinally strigose dorsally beneath
Wings; sternaulus broad, well impressed, with 8 or 9 weak but distinct cross
carinae; prepectal carina complete; prepectal area more or less smooth with
weak surface irregularity; propodeal carinae all developed, but rather deli-
cate; dorsal areas granular to weakly rugose-reticulate; basal carina a little
more than half as long as its forks; costulae more or less perpendicular to
forks; areola elongate, a little more than twice as long as broad, the sides

Table 4. Measurements and ratios of selected characters for females of the
pallidus species group. Top line in each instance gives range;
number below is for holotype or lectotype.

pallidus parvulus impus comes
n 10 13 17 10
number of 26-33 21-29 20-28 26-37
flagellomeres broken 21 24 36
body length 2.2-2.8 2.0-3.3 - 1-35.38 2.070. 1
(mm) 3.4 2.1 2.9 5. 0
FWL (mm) 1..9-3.2 1.8-2.8 1.5-2.9 2.3-3.9
broken 1,9 2.5 3.9
MS/EH 0.44-0.50 34-0. 58 43-0.58 0.47-0.59
0.50 0.35 0.48 0.54
TW/EH 0.54-0. 71 00-0. 63 50-0.63 0.63-0. 78
0.69 0,95 0.56 0.69
IOD/EH 1.03-1,21 96-1. 26 OO-1.19 .. 1. 09-1.28
todo 1, 05 1,12 1.14
OTW/OED 0.80-0. 93 71-0. 91 60-0.83 0.68-0.88
0.92 0.80 0.69 0. 88
OED/OOD 0. 70-0. 81 80-1. 00 85-1.11 0.64-0.82
0.73 0.91 0.93 0. 64
Pet. L/(2-3)L 1.40-1. 70 33-1.45 26-1.65 1.04-1.59
1. 60 1,38 30 1.30
ICu,/Ro 0, 93-1, 14 87-1.14 92-1.20 0.87-1.07
ee 1.08 0.94
OSL/FWL 0.60-0. 72 35-0. 64 D0-1,02. .0,91-1,57
- 0.39 0.68 1.26

58 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

sub-parallel, enclosing an area with several weak irregular transverse cari-
nae; fore tibia with about 15 spines arranged in irregular rows of 2, and a row
of 5 or 6 spines at apex; outer apical lobe of hind tibia with apparently 3 (Fig.
18) indistinct spines situated far forward; subdiscoideus not interstitial.

Abdomen: Petiole arched at base, longitudinally strigose-granular dorsally;
tergite (2-3) with a distinct area of weak granular or scaly-reticulate sculpture
on about basal 3/8, the remainder smooth; remaining tergites smooth, the sixth
tergite obliquely tilted ventrally as seen from the side, giving the gaster a
somewhat truncated apex in outline; ovipositor about 0.67x FWL.

Color: Body color wholly testaceous except gaster beyond basal half of
tergite (2-3) which is abruptly darker, light brown; eyes iridescent greenish-
blue in white light; ocelli dark brown; antennae testaceous at base, becoming
light brown apically; legs dark testaceous, the hind femora and tibiae slightly
darker than the corresponding segments of the fore and middle legs; tibiae
indistinctly lighter banded on about basal 1/4; ovipositor sheath testaceous
basally, becoming darker apically; fore wing infumated with two light brown
transverse bands separated by a hyaline band through the basal 1/3 of the stig-
ma, the apex also hyaline; veins and apical 2/3 of stigma light brown.

Variation: Paratype females. Little variation in sculpture intensity or
color occurs, except that occasionally the middle 1/3 of tergite (2-3) may show
faint sculpture traces; number of spines on the fore tibia varies from 12-26;
number of scutellar furrow cross carinae varies from about 4-8; spines on the
outer apical lobe of the hind tibia are sometimes inconspicuous and occasionally
apparently absent. Variation in other measurements and ratios given in Table

Paratype males: Essentially parallel the variation in the females, except
that the tarsi are comparatively shorter, about 3/4 as long as their respective
tibiae, and the sculptured area at the base of tergite (2-3) extends usually to
about the mid-point of the tergite.

Paratypes: 899, 40%c", same data as holotype [USNM]. North Carolina:
Tryon, 399, 400’ [no date], Hopkins No. 3634d (W. F. Fiske) [USNM]. South
Carolina: Yorkville, 19, 9 July 1913, Juniperus, Hopkins No. 9194q (J. E.
Smith) [USNM].

Remarks: The paratypes are all reared from red cedar (Juniperus virgini-
ana). In addition, I tentatively include in this species two series from Rye,
New Hampshire (Hopkins Nos. 3512 and 3513b) reared from northern white
cedar (Chamaecyparis thyoides) which agree with the paratypes in all charac-
ters except the ovipositor length which is distinctly longer and not overlapping
(0.94-1.02x FWL versus 0.50-0. 72x FWL in red cedar reared material). Also,
those from Chamaecyparis average slightly more flagellomeres (25-28 versus
20-26). Also, I include four females reared from arborvitae (Thuja occidental-
is) (Hopkins No. 3751), whose ovipositors are intermediate (OSL/FWL 0. 67-
0.83) and two short series with ovipositors about 0.50 FWL from bald cypress
(Taxodium distichum) (Hopkins Nos. 1239b and 3378d). In all other characters,
measurements and ratios, this material shows no distinctive differences, and
all are parasitic on Phloeosinus species infesting conifers belonging to the
families Cupressaceae and Taxodiaceae. Additional reared series from Thuja,
Chamaecyparis and Taxodium are needed to determine whether the differences
cited are real or representative of the extremes of a continuum.

The distinguishing characters of this species are the combination of the
relatively narrow vertex and temples (see Table 4); iridescent eyes; two well-
developed sub-parallel longitudinal carinae where the notauli converge; the
darker gaster beyond the yellowish basal half of tergite (2-3); and the presence

Matthews: Nearctic Spathius 59

of but 2-3 poorly developed (often inconspicuous) spines (Fig. 18) situated well
anteriorly on the outer apical lobe of the hind tibia. Spathius impus appears
most closely related to S. parvulus and S. pallidus.

The name is an arbitrary combination of letters.

Distribution: Probably widespread over the range of its host trees, in
eastern North America. In addition to the paratype localities, I have seen
specimens from Massachusetts, Michigan, New Hampshire, New York, Virgin-
ia, West Virginia, Florida, Louisiana, and Ottawa, Canada.

Hosts: I have seen several series from red cedar whose host is recorded
as either Phloeosinus sp. or P. dentatus (Say) (Scolytidae), under Hopkins Nos.
2590a, 2591a, 2723d, h, i, j andk, 2884f, 3634d, 3639b, 6136a, 6412, 6493a,
6954, and 10762a. Material from white cedar which I provisionally include in
this species (see Remarks) is recorded as reared from Phloeosinus sp. or P.
canadensis Swaine, under Hopkins Nos. 3512, 3513b, 3751, and 3751b. The
few specimens I have from bald cypress from southern is S. are reared from
Phloeosinus sp. under Hopkins Nos. 1239b and 3378d.

Hopkins’ (1892) record of Phloeosinus graniger (= dentatus) in red cedar as
a host of Spathius canadensis almost certainly belongs to this species, although
I was unable to locate the specimens, which according to his original manu-
script are labeled with number 2932. Also, Schedl’s (1932) and Beal and
Massey’s (1945) records of Spathius tomici parasitic on the above hosts (and
subsequent references to these records listed in Bushing, 1965) should be
changed also to this species, as I have seen their respective reared series.

15. Spathius comes, new species
(Fig. 31)

Holotype.- female, Wisconsin: Neopit, 7-31-39, Hopk. U.S. 32852n,
Lot No. 39- 18515 ex. Melanophila fulvoguttata (H. C. Secrest) TUSNM
Type No. 70505].

Description of type female: Body length, 5.0 mm; FWL, 3.9 mm; oviposi-
tor sheath length, 4.9 mm. Other measurements and ratios in Table 4.

Head: Flagellomeres 36; face strongly transversely strigose; frons with
strong transverse striae, whose extremities are gently curved ventrally; ver-
tex transversely strigose of uniform intensity from ocellar triangle to occipital
carina, and becoming granular laterally at eyes; temples posteriorly strigose
like vertex but becoming smooth along eye margin and at malar space; ocellar
triangle slightly raised and delicately transversely strigose.

Thorax: Lateral areas of prothorax weakly developed with irregular di-
agonal cross carinae merging posteriorly into horizontally wrinkled to granular
sculpture of remainder of propleuron; pronotum with a distinct shallow pit dor-
sally immediately anterior to the mesonotal margin; notauli shallowly crenu-
late anteriorly, converging into a broad, shallowly depressed rugose area;
mesonotum abruptly declivous to pronotum, the lobes evenly granular; scutellar
furrow about as broad as distance between lateral ocelli, and crossed by 7
variously developed carinae which converge slightly anteriorly; scutellum broad,
weakly convex, the disc evenly granular; mesopleural disc predominantly hori-
zontally finely striate-granular dorsally to granular above sternaulus; sternau-
lus relatively broad and shallow, crossed by about 17 delicate carinae which
fan out slightly dorsally and quickly fade into the granular sculpture, except
posteriormost elements which continue and gently arch posteriorly to meso-
coxal cavity; prepectal carina complete; prepectal area with surface irregu-

60 Contrib. Amer. Ent. Inst., vol. 4, no. 5, .1970

larity and numerous setae; basal carina of propodeum and its forks mostly ob-
scured by strong rugosity; costulae and carinae bounding area petiolaris dis-
tinct and well developed but with numerous irregular short branches; dorsal
areas coarsely granular to rugose; lateral faces of propodeum rugose; areola

» and area petiolaris rugose; fore tibia with 28 spines arranged in irregular rows

of two and with a row of 7 spines at apex; outer apical lobe of hind tibia with a
row of 3 spines situated anteriorly; hind coxae subglobular, coarsely granular
with weak oblique striation medially and dorsally; subdiscoideus not intersti-
tial; mediella with a very short third abscissa.

Abdomen: Petiole about as long as middle tibia, stout and moderately arched
basally; petiole sculpture dorsally predominantly rugose, becoming irregularly
longitudinally striated apically; epipleurite of petiole extending to about petiole
mid-point; tergite (2-3) with basal 13/32 weakly granular to scaly-reticulate,
the remainder of the tergite smooth and shining; lateral margin of tergite (2-3)
thickened on about basal 1/2; tergites 4 to 8 appearing smooth and shining, but
with very faint sparse punctulation anterior to setal rows revealed at high
magnification under proper light, this becoming progressively stronger apically,
tergites 7 and 8 quite obviously punctulate-granular; ovipositor broadly curved
and nearly as long as body.

Color: Head light orange-brown, the lower face and malar area weakly
suffused with dirty yellow; prothorax, mesonotum and scutellum light ferru-
gineous; propodeum and mesopleura dark ferrugineous; petiole basally dark
ferrugineous, apically suffused with bright orange-brown; gaster shining ferru-
gineous, tergites each with a lighter transverse subcuticular band anterior to
setal row and sternites heavily suffused with dirty yellow; trochanters, maxil-
lary and labial palpi and fore and mid coxae stramineous; antennae, tegulae and
hind coxae and remainder of fore and middle legs testaceous, except all tibiae
distinctly light banded on about basal 1/4; hind femora and tibiae light ferrugi-
neous; ovipositor sheaths ferrugineous to brown; fore wings infumated and dis-
tinctly banded as in Figure 31; basal third of stigma stramineous, remainder
brown; costal vein testaceous, remaining veins light tan.

Variation: Paratype females. Variation in gaster color ranges from wholly
ferrugineous to nearly castaneous, the subcuticular bands often obscure; basal
1/2 of tergite (2-3) is often lighter, suffused with bright orange-brown; also,
lower half of mesopleura (including sternaulus) may be lighter than remainder
of thorax; number of fore tibial spines varies from 17 to 28 with a row of 6-8
at apex; sculpture intensity of mesopleural disc, base of tergite (2-3) and ver-
tex is less strong on smaller specimens; propodeal carinae usually are partially
obscured as in type, but if apparently distinct, never thick and prominent; scu-
tellar furrow carinae variously developed, sometimes somewhat wrinkled and
irregular; variation in measurements and ratios given in Table 4.

Paratype males: Variation tends to parallel that of females except that mar-
gin of most gastral tergites is thickened laterally.

Paratypes: 729, 20%o’, same data as holotype except 8 emerged 22 Aug. 1939
and are labeled Hopk. No. 328520 [USNM]. Canada, Ontario: Waubamic,

299, 1o°, 2, 6& 7 July 1915 (H. S. Parish) [CU]. U. S. A., New Hampshire:
Durham, 499, 5o%o’, [May-June 1905] Hopk. U. S. 3516cl1 and 3516c2, Tsuga
bred (W. F. Fiske) [USNM]; Rye, 19, 10°, 1 July 1905, Hopk. U. S. 35128,
Chmcyprs thyoides bred (W. F. Fiske) [USNM]. Ohio: Mentor, 299, 1c, 15
May 1932, reared from hemlock borer (J. S. Houser) [USNM]. Pennsylvania:
Hummelstown, 299, Hopk. U. S. 15399c, ex. Melanophila fulvoguttata in
Tsuga canadensis (J. N. Knull) [USNM]. Wisconsin: Sawyer Co., 299, 4 &
30 June 1955, ex. Melanophila fulvoguttata (Hussain) [RDS].

Matthews: Nearctic Spathius 61

Remarks: This species superficially resembles Spathius simillimus Ash-
mead (see also ‘‘Remarks’’ under S. simillimus). It may be distinguished
from simillimus by the much longer ovipositor, absence of a well developed
punctulate band anterior to the transverse row of setae on tergites 4-6, and by
the narrower, less strongly granular vertex. Also, S. simillimus tends to
have all propodeal carinae very thick and distinct, which is never true inS.
comes.

Five females in the Porter collection from Hudson, Maryland, which differ
in their longer ovipositor (about 1.7x FWL), more flagellomeres (39-43) and
uniformly horizontally strigose mesopleural disc are tentatively included here
until additional material is available.

Distribution: Probably widespread throughout eastern North America. In
addition to the paratypes, I have seen specimens from Quebec, Nova Scotia,
New York, Maryland, and West Virginia.

Hosts: Most of the reared specimens before me were bred from the bupre-
stid Melanophila fulvoguttata (Harr.) in hemlock. Exceptions are the paratypes
from Rye, N. H. (Hopk. No. 3512f) labeled as reared from Chamaecyparis
thyoides (white cedar) from the scolytid Phloeosinus sp. This seems possibly
a labeling error, since a long series with the same labels but Hopkins Nos.
3512 and 3512b belong to S. impus sp. n. which is commonly parasitic on
Phloeosinus in red and white cedar.

Hopkins (1892, 1899) records the cerambycid Chrysobothris pusilla Cast.
in spruce as a host for Spathius simillimus; this should be changed to S. comes,
as the specimens (nos. 2557 and 2579, see ‘‘Remarks’’ for simillimus) proper-
ly belong to this new species.

16. Spathius benefactor, new species

Holotype.- female, New York: Lowr. Westchester Co., Reared elm sect.
22-27 May 1934 (A. P. Jacot) (letter 8-11-34) [USNM Type No. 70506].

Description of type female: Body length, 3.4 mm; FWL, 2.6 mm; oviposi-
tor sheath length, 2.25 mm. Other measurements and ratios in Table 6.

Head: Flagellomeres 25; face transversely strigose and somewhat shining,
becoming weakly rugose at malar spaces; frons delicately transversely striate;
vertex faintly transversely strigose and shining, becoming smooth at eyes;
temples posteriorly faintly wrinkled, more or less parallel to occipital carina,
becoming smooth at eyes and malar spaces; ocellar triangle weakly raised
and delicately strigose-granular; malar space slightly less than 1/2 eye height.

Thorax: Lateral areas of prothorax distinct anteriorly and narrow, with 3
or 4 oblique cross carinae, losing definition posteriorly in the predominantly
horizontally strigose sculpture of the propleura; notauli narrow and shallowly
crenulate anteriorly, meeting in a moderately depressed weakly rugose area
from which 2 short posterior carinae emerge which extend to the posterior
margin of mesonotum; mesonotum moderately declivous, the lobes evenly granu-
lar; scutellar furrow with 4 or 5 straight cross carinae; scutellum very weakly
convex, the disc evenly granular; mesopleural disc smooth and shining above
sternaulus, becoming weakly strigose posteriorly and longitudinally wrinkled
dorsally below wings; sternaulus moderately impressed with about 12 ill-defined
cross carinae, the most posterior elements apparently continuing and curving
posteriorly to mesocoxae; mesosternum smooth, shining; prepectal carina com-
plete; prepectal area smooth with weak irregular surface rugosity and sparse
short setae; propodeal carina distinct but somewhat irregular, strongest and

62 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

best defined on posterior half of propodeum posterior to the insertion of the
costulae; dorsal areas granular to weakly rugose in vicinity of basal forks and
costulae; basal carina delicate and irregular, shorter than its forks which are
also delicate and somewhat dissected with numerous short side branches;
costulae strongest and best defined at junction with basal forks, becoming dis-
sected and losing definition laterally; areola elongate, ill-defined anteriorly
and enclosing weaker, somewhat reticulate rugosity; fore tibia with 20 spines
arranged in an irregular single to double row and a row of 7-8 spines at apex;
outer apical lobe of hind tibia with a row of 5 or 6 spines at apex; hind coxae
sub-globular, granular to weakly obliquely striate dorsally; subdiscoideus vein
not interstitial.

Abdomen: Petiole strongly longitudinally strigose dorsally, becoming
somewhat reticulate rugose basally; tergite (2-3) with distinct area of granular
to scaly-reticulate sculpture on about basal 2/5 followed by a very narrow
smooth space and then very much weaker central scaly-reticulation ending be- —
fore the transverse row of setae, the apical margin smooth; lateral margin
of tergite (2-3) weakly thickened at base; remaining tergite with transverse
bands of sparse punctulation anterior to the transverse row of setae on each
tergite, the 7th and 8th tergite with a somewhat more strongly shagreened
sculpture, especially laterally; ovipositor 0.87x FWL.

Color: Body color basically shining ferrugineous, except the pronotum, the
lower half of the mesopleura and apex of the petiole to about middle of tergite
(2-3) lighter, suffused with rufo-testaceious; antennae testaceous, becoming
darker apically; maxillary and labial palpi stramineous; coxae and trochanters
light to dark testaceous, the hind coxae darkest; femora and tibiae shining
dark testaceous to ferrugineous, the hind darkest, the tibiae with a distinct
lighter band on about basal 1/4; tarsi typically testaceous, the last tarsomere
slightly darker; eyes light gray with an iridescent greenish hue in white light;
fore wings with two broad infumated bands separated by a hyaline transverse
band through the basal 1/3 of stigma; veins tan to stramineous.

Variation: Paratype females. Distinct transverse bands of sparse punctu-
lation on the gastral tergites occur in about half of the specimens examined,
although their visibility often depends largely on whether the tergites are suf-
ficiently distended to expose the area anterior to the setae; the middle area of
tergite (2-3) after the basal sculpture is typically smooth or at most weakly
scaly-reticulate as in type; the area of sculpture at the base of tergite (2-3) is
always present but sometimes very weakly developed; vertex sculpture varies
from apparently smooth to well developed transverse strigosity; the mesopleur-
al disc is nearly always smooth immediately above sternaulus but sometimes
scaly-reticulate; sternaulus is typically expressed as in the type but sometimes
is very shallow and obscure, and rarely is sharply infolded as a deep groove;
the form of the propodeal carinae is typically as in type, but the basal carina
and its forks vary from sharply defined to almost completely obscured; body
color varies from dark testaceous to dark ferrugineous, but is typically shin-
ing light ferrugineous with the apex of the petiole to about the middle of tergite
(2-3) lighter rufo-testaceous; the lower half of the mesopleura and the prono-
tum may or may not be lighter than the remainder of thorax; number of fore
tibial spines varies from 12-30 with a row of 5-9 spines at the apex; the outer
apical lobe of the hind tibia has from 3-6 spines but typically 4 or 5; variation
in other measurements and ratios is given in Table 6.

Paratype males: Essentially parallel the same sorts of variation in sculp-
ture and color noted for the females, except that area of sculpture on base of
tergite (2-3) typically covers about basal 1/2 of tergite, the lateral margins of

Matthews: Nearctic Spathius 63

at least the anterior tergites are thickened, and the areola is usually slightly
smaller than on the females, with the basal carina thus often longer than its
forks.

Paratypes: 599, 9o’o', same data as holotype [USNM, CU]. Canada:
Quebec: Berthierville. 999. 20’c’, 6 Aug. 1946 (A. Robert) [Montreal, CNC].
Massachusetts: Berkshire Co., Williamstown, 799, 800", 1 July 1965 [MCZ].

Remarks: Since Kaston and Becker’s (1936) paper, the name Spathius cana-
densis Ashmead has come to be applied almost automatically to any Spathius
reared from elm. However, detailed study of Ashmead’s type (a female from
Ottawa, Canada) and over 45 different series of Spathius from elm (resulting
largely from the emphasis on Dutch Elm Disease research beginning in the
mid-1930’s) has convinced me that elm-reared Spathius belong to a single,
undescribed species, and that S. canadensis is actually a senior synonym for
S. claripennis and S. tomici. (See additional comments under the ‘‘Remarks’’
section of canadensis.) The type of S. canadensis, while having many simil-
arities to benefactor in ‘‘gestalt’’, differs significantly in many respects, most
notably by the type’s relatively short ovipositor (0.54x FWL versus 0. 62-1. 09x
FWL for benefactor), but also in its slightly greater ratios of IOD/EW and TW/
EH and smaller ratio of OTW/OED.

In addition to the characters given in the key, Spathius benefactor is dis-
tinguished by the short malar space, always less than half of eye height, the
typically wrinkled basal carina, greenish iridescent eyes and the typically
smooth mesopleural disc above the sternaulus. Larger individuals are super-
ficially similar to S. laflammei, but usually the combination of fewer flagel-
lomeres (21-32 versus 29-42), fewer spines on the outer apical lobe of hind
tibia (3-6, typically 4 or 5, versus 5-8 in laflammei, typically 6 or 7) and the
narrower vertex of benefactor (see Table 6) will distinguish the two.

Biology: Under the name canadensis, this species has been investigated in
some detail, particularly by Kaston (1937, 1939) and Robert (1951a & b, 1961);
the life stages are figured by Kaston (1937, 1939) and Robert (1961). Kaston’s
drawing is reproduced in Clausen (1940) and Craighead (1950). Snodgrass
(1941) has figured the male genitalia.

Kaston has studied the biology of this species’ parasitism of the elm bark
beetle Hylurgopinus rufipes, and Robert discusses it on three species of wee-
vils (Magdalis barbita, M. armicollis and M. inconspicua) based on large samp-
les over a four year period. The work of both can be summarized as follows:
Spathius benefactor has two, perhaps as many as four, generations per year,

a complete cycle requiring 5-6 weeks; at least two generations can be reared
on the same population of weevils. It overwinters in the prepupal stage, trans-
forming to the pupa in early May, the adults eclosing in late May. The sex
ratio in the Quebec populations (Robert, 1960) was weighted in favor of the
males, females never comprising more than 30-40% of the populations. Kas-
ton states that this species attacks only the ultimate and penultimate instars,
usually parasitizing only 5-10% (seldom more than 25%) of the beetle popula-
tion. Robert records average levels of parasitism of 7.7-15.3% over the
period 1947-1950, the highest recorded level being 57.7% at one locality in
1948.

Distribution: Throughout eastern North America north of North Carolina.
I have seen several hundred specimens, representing Connecticut, Illinois,
Kansas, Maryland, Massachusetts, Michigan, Minnesota, Missouri, New
Hampshire, New Jersey, New York, North Carolina, Ohio, Pennsylvania,
Vermont, Virginia, Wisconsin, and Ottawa and Quebec.

Hosts: Of the hosts formerly recorded from S. canadensis (see Bushing,

64 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

1965), the following should be transferred to this species: Hylurgopinus rufipes,
Scolytus multistriatus, S. rugulosus, Magdalis barbita, M. armicollis and ©
Leperisinus aculeatus. (For the status of the others, see under ‘‘Hosts’’ of
Spathius canadensis as herein defined.) The only additional record is Magdalis
inconspicua Horn (Robert, 1960) from elm.

17. Spathius laflammei Proffancher
(Fig...23)

Spathius laflammei Provancher, 1880, Naturliste can. 12: 164, fig. 12. [Type
locality: Unknown, probably near Quebec City, Canada (lectotype female
in Laval Univ. , Quebec).

Females: Body length, 3.5-6.1 mm; FWL, 3.0-4.7 mm; ovipositor sheath
length, 2.0-3.9 mm. Other measurements and ratios in Table 6.

Head: Flagellomeres 29-42; face coarsely transversely strigose; frons with
strong transverse striation; vertex typically with weak transverse striation an-
teriorly behind ocelli, this quickly diminishing in intensity, the posterior half
smooth and shining; temples nearly smooth, with weak rugosity posteriorly
paralleling the occipital carina; ocellar triangle slightly raised and finely granu-
lar.

Thorax: Lateral areas of prothorax moderately developed with several
oblique cross carinae, becoming obscured posteriorly in strong longitudinal
strigosity of remainder of prothorax; notauli deep and crenulate anteriorly,
meeting in a broad moderately depressed rugose area; mesonotum robust and
abruptly declivous, the lobes evenly granular; scutellar furrow with 4-9 well
developed straight cross carinae; scutellum broad, somewhat projecting, the
disc granular, becoming weakly rugose at posterior margin; mesopleural disc
typically smooth and shining above sternaulus (occasionally weakly scaly-
reticulate), abruptly becoming longitudinally rugose dorsally and granular
posteriorly; sternaulus well impressed, broad with several distinct cross
carinae, the posterior elements continuing and curving posteriorly to meso-
coxal cavities; prepectal carina complete; prepectal area more or less smooth;
propodeal carinae often irregular or obscured in reticulate rugosity anteriorly,
best defined on posterior portion of propodeum; dorsal areas coarsely granular
to reticulate-rugose; basal carina obscured to irregularly wrinkled, slightly
shorter than its forks which are also irregular and dissected with numerous
short reticulate side branches; areola longer than broad, often poorly defined
anteriorly and usually enclosing weak reticulate rugosity; fore tibia with 24-47
spines arranged in irregular row of two and occasionally three, and a row of
7-11 spines at apex; outer apical lobe of hind tibia with a row of 5-8 conspicu-
ous spines (Fig. 23), typically 6 or 7; subdiscoideus vein not interstitial.

Abdomen: Petiole stout, reticulate rugose to longitudinally strigose dor-
sally; tergite (2-3) with distinct area of granular sculpture on about basal 1/3
to 2/5, the remainder of the tergite usually smooth, shining, but occasionally
with faint sculpture on middle of tergite anterior to the transverse row of
setae; remaining tergites each typically with a well developed transverse
punctulate band anterior to the setal row, this sometimes obscured or covered
by the overlapping margin of the previous tergite, the 7th tergite often sha-
greened dorsally; ovipositor 0. 61-0. 89x FWL.

Color: Body color typically ferrugineous, but varying from dark testaceous
to nearly piceous, the head and apex of petiole to basal half of tergite (2-3)

Matthews: Nearctic Spathius 65

lighter and the gaster beyond the basal half of tergite (2-3) darker, especially
on darker individuals; coxae, trochanters and tarsi light testaceous, the hind
coxae darker; femora and tibiae dark testaceous, usually nearly concolorous
with thorax, the hind legs slightly darker; antennae dark testaceous, becoming
darker apically; eyes gray, sometimes with a greenish iridescent hue in white
light; ovipositor sheaths testaceous at base, becoming dark brown apically;
fore wings with two broad dark transverse bands separated by a hyaline band
through the base of stigma.

Males: Essentially agree with the females but often much smaller (2. 0-
4.5 mm), with as few as 20 flagellomeres, and the vertex is usually completely
smooth and shining; sculpture on the base of tergite (2-3) usually covers about
the basal half of the tergite and lateral margins of the tergites are thickened.

Remarks: Spathius laflammei, described in 1880, has always been poorly
known, and no host records are reported. Although Ashmead (1893) included
S. laflammei in his key, his concept of the species was based on something
other than the type, and instead his characterization of it agrees exactly with
the new species, S. stigmatus (see ‘‘Remarks’’ under that species). Through
the courtesy of Dr. René Béique of Laval University, Quebec, I have been able
to examine Provancher’s type and compare it with other specimens. The type
is a pinned female and bears a yellow label with the printed number 563; be~
neath that is a handwritten label ‘‘Spathius laflammei Prov.’’ followed by a
red lectotype label by P. M. Marsh, Feb. 1965. The fore tarsi, both middle
legs and the right fore wing of the lectotype are missing; the remaining wings
are very dirty and folded together. (For additional notes on Provancher’s
type, see Gahan and Rohwer, 1918, and Shenefelt, 1968.)

I tentatively include here several very flattened individuals which at first
look quite different, their profile resembling that of S. calligaster (Fig. 17).
They agree with the typical S. laflammei in sharing the broad, nearly smooth
vertex and temples, but differ in the thorax shape, with the mesopleural disc
usually scaly-reticulate to granular, and the body size is smaller (ranging up-
wards from 2.8 mm), with correspondingly fewer flagellomeres and shorter
Wings and ovipositor. All are inthe U. S. National Museum collection,
reared from hickory under Hopkins Nos. 1418a, 3649f, 3653e & f, and 10938c.

Diagnosis: This is a relatively large and robust species apparently related
to S. simillimus and S. floridanus, sharing the relatively broad vertex and
temples. It may be distinguished from them and from the other Nearctic spe-
cies by the smooth mesopleural disc immediately above the sternaulus, almost
smooth vertex and temples, and the greater number of spines on the outer api-
cal lobe of the hind tibia (Fig. 23) (6-8 spines, versus 4-6 on simillimus and
2-4 on floridanus).

Distribution: Apparently widespread through eastern North America south
to North Carolina. I have seen about fifty specimens, representing Connecti-
cut, Kansas, Maryland, Massachusetts, Michigan, New Hampshire, New Jer-
sey, New York, North Carolina, Rhode Island, Virginia, Wisconsin and Quebec,
Canada.

Hosts: No hosts were previously recorded. I have seen several specimens
reared from hickory (Carya spp.), all of which, where known, are reported as
parasitic on the weevil Magdalis olyra under the following Hopkins numbers:
3262c, 3528xlb, 3605d, 3649a, 3654d and 10761ly. Also I have seen a single
male labeled ‘‘ex Eucrada humeralis’’ (Anobiidae), and a series labeled ‘‘bred
from beech’? (AMNH).

66 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

18. Spathius trifasciatus Riley
(Figs. 15, 29)

Spathius trifasciatus Riley 1873, Rep. Insects No. 5: 106. [Type locality:
Missouri: St. Louis (C. V. Riley) (allotype male in ANSP - not USNM
Type No. 2764, see Remarks)].

Spathius unifasciatus Ashmead 1893, Can. Ent. 25: 70, 72. [Teste Muesebeck
and Walkley, 1951, Agriculture Monogr. 2: 170. Type locality: West
Virginia: Morgantown (holotype female in USNM, Type No. 2149)].

Females: Body length, 3.0-6.5 mm; FWL, 2.3-4.4 mm; ovipositor sheath
length, 2.0-4.9 mm. Other measurements and ratios in Table 5.

Head: Antennae about as long as body; flagellomeres 25-36; face trans-
versely strigose-granular; frons transversely striate, the central portion be-
coming irregularly rugose in large individuals; ocellar triangle usually distinct-
ly raised and transversely striate; vertex delicately striate anteriorly, the in-
tensity quickly diminishing posteriorly to become smooth and shining at occipi-
tal carina; temples delicately strigose more or less parallel to occipital carina,
the intensity strongest adjacent to occipital carina and diminishing anteriorly
toward eye and ventrally toward malar space which is smooth.

Thorax: Prothoracic lateral areas poorly defined, shallow, weakly crenu-
late anteriorly, merging posteriorly into the strongly developed predominantly
horizontal strigose sculpture covering remainder of propleura; notauli deep and
crenulate anteriorly, converging into a strongly depressed rugose area from
which usually emerge two or three distinct short carinae which reach posterior
margin of mesoscutum; mesonotum broad, moderately declivous, the lobes
evenly granular; scutellar furrow deep, crossed by 5-8 strong carinae; scu-
tellum moderately convex (but usually less so than in S. brunneus), the disc
evenly granular and sides strongly obliquely striate; mesopleural disc longitu-
dinally strigose to granular in smaller individuals, dorsally the striae become
progressively stronger to reticulate-rugose on the anterior-dorsal region;
sternaulus depressed, bordered below by a well developed gently curved longi-
tudinal carina extending the length of the mesopleuron from the prepectal carina
to the mesocoxal cavity and with several irregular oblique cross carinae above
which curve dorsally and posteriorly; prepectal carina complete, strong; pre-
pectal area sparsely punctate with hairs to apparently smooth; propodeal cari-
nae development variable, the carinae usually partially obscured by the reticu-
late-rugose sculpture which predominates; basal carina usually much shorter
than its forks; areola and area petiolaris usually confluent and slightly concave,
more or less dominated by reticulate-rugose sculpture; dorsal areas granular
basally, becoming reticulate-rugose in region of costulae and basal forks;
fore tibia with 27-38 spines arranged in irregular rows of 2 or 3 and a row of
6-9 at apex; outer apical lobe of hind tibia with a row of 6-10 spines; hind coxae
granular to obliquely striate dorsally and angled basally below; tegulae mostly
finely granular; subdiscoideus not interstitial, arising above middle of brachial
cell; first intercubitus almost always distinctly longer than second radial ab-
scissa.

Abdomen: Petiole dorsally longitudinally strigose at apex to reticulate-
rugose basally; tergite (2-3) usually with about basal 1/3 to 1/2 delicately scaly-
reticulate, the remainder smooth or with punctulation laterally anterior to setae
row (occasionally punctulation extends completely across tergite as a band in
very large specimens); remaining tergites with narrow transverse punctulate

Matthews: Nearctic Spathius 67

to weakly granular bands anterior to the setae row, these bands variously
developed, but strongest laterally, the central portions sometimes obscured
and often not visible due to the overlapping margin of previous tergite; epi-
pleurites weakly punctulate; gastral sculpture intensity seems to vary according
to individual size, smaller specimens having gasters which appear nearly
smooth; margin of tergite (2-3) weakly thickened at base.

Table 5. Measurements and ratios of selected characters for females of the
trifasciatus species group. Top line in each instance gives range;
number below brunneus is for lectotype; trifasciatus holotype lost.

trifasciatus brunneus
n 13 11
number of 25-36 31-34
flagellomeres broken
body length (mm) 3.0-6.5 2.9-4.9
4.5
FWL (mm) 2.3-4.4 2.7-3.8
3.6
MS/EH 0,44-0.53 0.43-0. 55
0.49
TW/EH 0.64-0.95 0.69-0. 81
0. 74
IOD/EH 1.02-1. 24 1.00-1.14
1.09
OTW/OED 0.72-1.00 0.67-0. 82
0. 74
OED/OOD 0.60-0. 78 0.59-0. 70
0.63
Pet. L/(2-3)L 1.11-1.61 135-4, 65
Ue IS)
ICu,/Ro 0.92-1.42 0.97-1. 04
bes 1, 03
OSL/FWL 0.87-1.20 0.45-0. 68
0.50

Color: Body predominantly typically dark ferrugineous to piceous, except
about apical half of petiole and about basal half of tergite (2-3), which are near-
ly always distinctly lighter, usually fulvous, light ferrugineous or yellow; head
usually somewhat lighter than thorax; thorax with propodeum darkest, sometimes
(along with base of petiole) nearly black, the prothorax, prepectal areas and
lower mesopleura (sternauli) usually slightly lighter than dorsal parts (very

68 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

much lighter, stramineous, in one female from Mississippi); eyes usually an
iridescent bluish-green in white light; flagellomeres dull ferrugineous to near-
ly brown; ovipositor sheaths brown, though slightly lighter basally; palpi, fore
and mid coxae testaceous; hind coxae and all femora and tibiae dull ferru-
gineous, the hind darkest; tibiae with a distinctly lighter band on about basal
1/5; fore wing distinctly banded as in Figure 29; basal third of stigma pale
whitish yellow; veins pale brown to dull ferrugineous; tegulae light ferrugine-
ous to testaceous.

Males: Agree with females in most morphological features except femora
are much more swollen (Fig. 15) and margin of tergite (2-3) and most subse-
quent tergites is thickened laterally; also, number of spines on the fore tibia
is fewer, ranging from 10-24 arranged in an irregular single to double row.

Remarks: In the U. S. National Museum are four pins of supposed S.
trifasciatus Riley with red ‘‘USNM type no. 2769’’ labels catalogued by Ash-
mead in 1896: 299, 1c’, and 3 cocoons. In actuality, the adults represent not
one but three species of Spathius: S. trifasciatus (co), S. stigmatus, sp. n. (9),
and S. floridanus Ashmead (9). Noting this error by Ashmead, Rohwer later
affixed handwritten ‘‘Not Type. Roh.’’ labels on the two females and ‘‘Spathius
trifasciatus Riley Allotype male ? Roh.’’ to the male specimen. However, it
now appears that even this male is not the true allotype.

The source of Ashmead’s confusion may possibly have been the three co-
coons glued to a single point with the labels ‘‘21701”’ and ‘‘From Scolytus
caryae’’ in Riley’s handwriting. On file in the U. S. National Museum on 3x5
cards are excerpts from notes and letters of Riley; one card headed ‘‘No. 217R
Scolytus caryae Riley’’ states ‘‘Bred a Bracon from it - See 5th Rept. Marked
2170 in Box 2, also a second one 21791 in Box 2. (Both sent to Cresson Mch
20/73) 2 I suspect that Ashmead somehow obtained the cocoons, discovered
that the number matched that in Riley’s notes, and proceeded to catalogue
three specimens which he thought to have emerged from these cocoons without
checking them further.

However, in 1873 Cresson was at the Philadelphia Academy, and among
the Spathius which Dr. P. M. Marsh recently found there was a single male
S. trifasciatus labeled ‘‘Mo.”’ and ‘2179’ with a third label ‘‘Par. Scolytus
caryae’’ in Riley’s handwriting. I have no doubt that this is the true allotype
male. The female which should bear the label ‘‘21791” is apparently lost; it
is definitely not in the U. S. National Museum and Dr. Marsh, in a subsequent
trip to Philadelphia, has specifically searched for it, at my request, without
success.

Still unexplained are how Ashmead came to have the cocoons and mis-
matched adults and why the cocoons were not mentioned by Riley in either his
notes or original description. However, since Riley’s description is based
chiefly on the female and his original female specimen seems to be lost, one
alternative would be to select a neotype female in addition to the positively
identified allotype to henceforth stand as the name bearer for the species.
Since there is always the possibility that the lost type female may eventually
be found, I have decided not to select a neotype, but rather have labeled afe-
male from Topeka, Kansas (USNM), as plesiotype so that subsequent workers
will know the basis for my concept of S. trifasciatus Riley.

Additional evidence that Ashmead was confused as to the identity of Riley’s
S. trifasciatus is the fact that he described S. unifasciatus which has proved to
be synonymous (teste Muesebeck and Walkley, 1951). Ihave seen Ashmead’s
type and agree that it is a synonym of S. trifasciatus. The type pin (USNM
Type No. 2149) has both a male and female, though Ashmead mentions only

Matthews: Nearctic Spathius 69

the female; however, the label number 3164 agrees with the Hopkins original
manuscript for his 1892 paper. Interestingly, in his table of species, Ash-
mead separates 8S. trifasciatus and S. unifasciatus in two different parts of the
key and states S. trifasciatus to have an ovipositor twice as long as its body,
which is clearly anerror. Unfortunately, this is but another indication of the
haste with which he often worked.

Diagnosis: Spathius trifasciatus may be readily distinguished from its
closest relative, S. brunneus, by the longer ovipositor, darker body color
and the strongly depressed rugose area where the notauli converge.

Distribution: I have seen well over 100 females and at least 50 males,
representing the following localities: Kansas, Maryland, Mississippi, Mis-
souri, New Jersey, New York, North Carolina, Pennsylvania, Texas, Virgin-
ia, West Virginia and the District of Columbia. These suggest that the range
of S. trifasciatus is eastern United States, south of a line from Massachusetts
to Kansas and west to Kansas and Texas, except the extreme southeast. Prob-
ably the range will be found to mirror that of its host trees, hickory (Carya
spp.), which occur throughout the deciduous forests of the eastern U. S.

Hosts: The first and most frequently recorded host in the literature is
Scolytus quadrispinosus Say (= S. caryae Riley) in hickory. Numerous authors
have cited this host and a summary of these may be found in Bushing (1965).

In addition, Blackman and Stage (1924, p. 165-166) list Magdalis olyra in
hickory as a probable host, and Felt (1915, p. 29) reports it parasitic on the
hemlock borer, Melanophila fulvoguttata Harr. In both cases, the evidence
suggests that the parasite wasps were misidentified.

In the case of Magdalis olyra, I have already cited the agreement between
Blackman’s specimen labels and his comments under 8S. stigmatus sp. n. In
addition, Blackman and Stage state (p. 166) that ‘‘In not a single case was it
[S. trifasciatus] found associated with E. quadrispinosus although the hickory
bark beetle had emerged two years previously from the material in several of
these cages.’’ I can further add that no true 8. trifasciatus were found among
the material borrowed from the Syracuse University College of Forestry col-
lection. The fact that one of the types catalogued as S. trifasciatus was S.
stigmatus (See Remarks) suggests the possible source of error.

In the case of the Melanophila host record, I have not seen the reared speci-
mens in question. However, the new species, S. comes, is reared from M.
fulvoguttata, and in many respects superficially resembles S. trifasciatus.
Also, none of the S. trifasciatus specimens before me are reared from hem-
lock. Therefore, Iam inclined to doubt the validity of this host record.

The only other published host record is Scolytus rugulosus (Ratzeburg)
(Chamberlin, 1939). Since I have no specimens which confirm this record, I
am inclined to question its validity as well, especially since it is an introduced
species and works primarily in Prunus, Pyrus and Crataegus, which are not
particularly closely allied to hickory.

The available evidence suggests that S. trifasciatus is largely restricted to
bark beetles (primarily S. quadrispinosus) and other borers in the various spe-
cies of hickory, as originally reported by Riley and by Ashmead for S. uni-
fasciatus. All reared specimens I have seen are from hickory (rarely oak),
many recorded with associated insects in the Hopkins’ card file at the U. S.
National Museum under the following numbers: 3262c, 3458c, 6850c, 9423b3,
10764-c, 10935i, 10936b, 10937b, 10938c, 10938d. In addition I have seen the
S. unifasciatus specimens referred to by Beal and Massey (1945) (from Durham,
North Carolina, no. 1573) and can confirm their identity. Their additional
reference to S. unifasciatus parasitizing Hylurgopinus rufipes is to be doubted.

70 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

19. Spathius brunneus Ashmead
(Fig. 14)

Spathius brunneus Ashmead 1893, Can. Ent. 25: 72. [Type locality: West
Virginia: Morgantown (A. D. Hopkins) (lectotype female in USNM,
Type No. 70507)].

Spathius bruneus [sic] Dalla Torre 1898, Cat. Hym. 4: 243.

Females: Body length, 2.9-4.9 mm; FWL, 2.7-3.8 mm; ovipositor sheath
length, 1.4-2.6 mm. Other measurements and ratios in Table 5.

Head: Antennae about 1.5x FWL; flagellomeres 31-34; face transversely
strigose-granular; frons finely transversely striate, becoming finely granular
at eyes; vertex delicately transversely striate to nearly smooth, especially
posteriorly adjacent to occipital carina; temples with delicate strigosity parallel
to occipital carina, becoming finely granular toward eyes and smooth at malar
space; ocellar triangle faintly granular to delicately transversely striate.

Thorax: Robust, about 0. 7x as high as long (Fig. 14); lateral areas of pro-
thorax shallow, narrow and smooth with about 9-12 oblique cross carinae,
these losing definition posteriorly to become obscured by the horizontally stri-
gose granular sculpture covering remainder of propleura; notauli shallowly
crenulate anteriorly converging to a slightly depressed rugose area from which
emerge usually two distinct carinae reaching posterior margin of mesonotum;
mesonotum moderately declivous, broad, the lobes evenly granular; scutellar
furrow broad, gently arched around anterior margin of scutellum and having
6-10 well developed evenly spaced cross carinae; scutellum strongly convex
(Fig. 14), the disc evenly granular and sides obliquely striate; mesopleural
disc centrally usually nearly smooth, becoming progressively granular, then
with delicately swirled striations at outer limits of area and dorsally longitu-
dinally strigose; sternaulus (Fig. 14) indistinct, smooth, and lacking cross
carinae, instead bordered below by a strong gently curved longitudinal carina
extending the length of the mesopleuron from the prepectal carina to the meso-
coxal cavity, with up to 5 shorter, variously developed carinae more or less
paralleling it above; prepectal carina complete, strong; prepectal area nearly
smooth, sometimes with a few irregular oblique carinae extending from the
anterior margin; propodeal carinae development variable, strongest posteriorly;
basal carina weakly developed, usually shorter than its forks; areola and area
petiolaris confluent, granular, crossed by several irregular carinae; costulae
more or less perpendicular to basal forks; dorsal areas granular basally, be-
coming reticulate-rugose at costulae; fore tibia with 35-47 spines arranged in
irregular rows of 2 or 3, anda row of 7-8 at apex; outer apical lobe of hind
tibia with a row of 6-7 spines; hind coxae basally angled below, with indistinct
oblique striations or weak granular sculpture dorsally; subdiscoideus not inter-
stitial, arising above middle of brachial cell; tegulae basally faintly granular,
apically smooth. :

Abdomen: Petiole dorsally uniformly longitudinally striate at apex, be-
coming irregularly longitudinally strigose to reticulate-rugose at base; ter-
gite (2-3) usually with a very weakly scaly-reticulate area on about basal 5/12,
followed by a narrow smooth space and then by a narrow transverse band of
sparse fine punctures immediately anterior to the subapical transverse row of
elongate setae, the tergite smooth posterior to the setae row; tergites 4-7 each
with similar transverse narrow punctulate bands just anterior to the setae row,
these bands usually mostly covered by the overlapping margin of previous ter-

Matthews: Nearctic Spathius 71

gite; ultimate tergite, when visible, appearing granular apico-laterally; varia-
tion in gastral sculpture often seems correlated with specimen size, i.e.,
larger individuals show much more strongly developed punctulate bands and
basal aciculation on tergite (2-3), while on smaller individuals the sculpture is
often very faint, the gastral tergites appearing smooth; margin of tergite (2-3)
weakly thickened at base.

Color: Body more or less uniformly testaceous except gaster beyond basal
half of tergite (2-3), which varies from dull ferrugineous to castaneous; palpi
stramineous to testaceous; antennae honey yellow to testaceous, the apical 10-
15 flagellomeres becoming dusky brown; flagellomeres each with a very narrow
darker apical band; eyes usually an iridescent bluish green in white light; man-
dible tips ferrugineous; legs testaceous, the tibiae, especially hind tibiae,
sometimes slightly darker and always distinctly lighter on about basal 1/5,
appearing banded; ovipositor sheaths testaceous basally, apical half brown;
fore wing distinctly banded very similarly to S. trifasciatus (Fig. 29); basal
third of stigma and sometimes costa pale yellow, other veins usually pale
brown; costula of hind wing usually pale yellow especially in area of hamuli;

tegulae pale yellow to testaceous. _
3 Males: Agree with females in color and most morphological features ex-
cept femora are much more swollen and margin of tergite (2-3) and most sub-
sequent tergites is thickened laterally; also gastral sculpture is usually reduced
in intensity, the punctulate bands often completely obscured and the basal half
of tergite (2-3) usually appearing smooth. Number of fore tibial spines is also
reduced, ranging from about 24-30.

Remarks: As was true for several of the Spathius described in his 1893
paper, Ashmead gave no formal description of S. brunneus beyond that in his
‘‘Tables of Species’’. The specimen which surely represents the type bears the
labels ‘‘179 Hopk. W. Va.’’ and ‘‘Spathius brunneus Ashm. o, 2, Ashm.’’ in
Ashmead’s handwriting. A third label inS. A. Rohwer’s handwriting says
‘¢Ashmead Type’’. Although Ashmead does not give a host record, the number
179 matches a number on Hopkins’ original manuscript for his 1892 paper in
which S. brunneus is reported parasitic on ‘‘Scolytus muticus (?) and Agrilus
fallax’’. On the pin are two females and one male, each on a separate point.

I have placed a red ink mark near the base of the point of the female having both
flagella missing and here designate this female as the lectotype. The other fe-
male and male thus become paralectotypes. Since S. brunneus was not previ-
ously catalogued in the U. S. National Museum type collection, it has been
designated USNM Type no. 70507.

Diagnosis: This species may be readily distinguished from its closest rela-
tive, S. trifasciatus Riley, by its lighter body color, shorter ovipositor, less
strongly depressed notauli, more conical scutellum (Fig. 14), and lack of cross
carinae on the sternauli.

Distribution: In addition to the type I have seen 12 females and 6 males, all
in the U. S. National Museum collection. All are reared from Celtis occiden-
talis and represent three localities: Plummers Island, Maryland; Mariana,
Florida; and Morgantown, West Virginia, the type locality.

Hosts: Three hosts are recorded: Scolytus muticus Say, Agrilus fallax
Say and Scolytus ventralis LeConte. The first two species were originally
reported as hosts by Hopkins (1892), and one or both repeated in Hopkins (1893a
and 1893b), Dalla Torre (1898), Pierce (1908), Chamberlin (1939) and Bushing
(1965). Scolytus ventralis is apparently a copy error introduced by Chamberlin
(1939) and repeated in Bushing (1965). Chamberlin cites Bedard (1938) as his
source, but Bedard referred nottoS. brunneus but to Spathius brunneriViereck, a

72 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

known parasite of Scolytus ventralis. No additional host records have been
found.

Table 6. Measurements and ratios of selected characters for females of the
laflammei and simillimus species groups. Top line in each instance
gives range; number below is for holotype or lectotype.

laflammei benefactor simillimus floridanus
n 10 23 12 14
number of 29-42 21-32 27-33 23-39
flagellomeres broken 25 27 broken
body length 3.5-6.1 2.2-4.3 2.9-4.5 1,.4-4.6
(mm) 4.7 3.4 a 4.0
FWL (mm) 3.0-4. 7 1.9-3.4 2.5-3.4 1.9-4.2
3.8 2.6 2.8 4.1
MS/EH 0.48-0. 63 0.41-0. 50 0.44-0.50 0.47-0. 67
0. 63 0.46 0.50 0.48
TW/EH 0.67-0. 77 0.55-0. 74 0.57-0. 67 0. 64-0. 85
0.77 0.57 0.62 0.67
IOD/EH 1.06-1. 33 0.96-1.12 1.00-1.17 1.00-1.33
1, 33 1.04 1odt 1.07
OTW/OED 0. 72-0. 89 0.75-1.07 0.69-0. 86 0. 64-0. 88
0. 73 0.86 Oy T1 0. 84
OED/OOD 0.58-0. 78 0. 71-0. 88 0.61-0. 73 0.56-0. 78
0. 73 0.78 0.70 0.59
Pet. L/(2-3)L 1.36-1.77 1.39-2.00 1.25-1. 70 1.10-1.61
£52 1,00 1.42 1.47
ICu,/R, 0.90-1.20 0.91-1.25 0.84-1.25 0. 88-1. 35
1.14 1.04 i eK 1.00
OSL/FWL 0.61-0. 89 0.62-1.09 0.58-0. 78 0.65-1.10
0.66 0.87 0. 64 0.98

20. Spathius simillimus Ashmead
Spathius simillimus Ashmead, 1893, Can. Ent. 25: 70, 73. [Type locality:
West Virginia: Morgantown (lectotype female in USNM, Type No.
2150)].

Females: Body length, 2.9-4.5 mm; FWL, 2.5-3.4 mm; ovipositor sheath

Matthews: Nearctic Spathius 73

length, 1.6-2.4 mm. Other measurements and ratios in Table 6.

Head: Flagellomeres 27-33; face coarsely granular; frons strongly trans-
versely strigose medially, giving way to granular sculpture at eyes and ocelli
level; vertex relatively broad, OOD distinctly greater than OED and TW, ver-
tex prominently and completely uniformly granular with only weak indication of
transverse strigosity, this extending to frons, eyes, and continuing onto temples
nearly to malar space; malar space not more than half of EH; ocellar triangle
granular, weakly raised. :

Thorax: Lateral areas of prothorax indistinctly defined, more or less ob-
scured posteriorly by the generally rugose to coarsely granular sculpture of
remainder of prothorax; notauli strong, crenulate anteriorly, meeting ina
broad slightly depressed area of strong rugosity; mesonotum abruptly declivous,
the lobes uniformly granular; scutellar furrow narrowest at mid-point, broad-
ening laterally and crossed by 6-9 variously developed, often wrinkled irregu-
lar carinae; scutellum weakly conical, the disc granular, becoming weakly ru-
gose at posterior margin; mesopleural disc evenly granular to scaly reticulate
immediately above narrow sternaulus, giving way to stronger longitudinal
strigosity dorsally; mesosternum below sternaulus also evenly granular to
scaly-reticulate; sternaulus a narrow moderately impressed furrow, very
slightly broader posteriorly and crossed by several variously developed, usual-
ly irregular carinae which end abruptly above, never continuing posteriorly or
onto mesopleural disc; prepectal carina complete; prepectal area more or less
smooth with weak surface irregularity and sparse setae; propodeal carinae
prominent, very thick and always well developed, especially the basal carina
and its forks; dorsal areas coarsely granular to weakly rugose; basal carina
usually slightly shorter than its forks; costulae typically distinctly forked al-
most at base, the posterior branch extending to the posterior-lateral angles
which are usually slightly projecting; areola an elongate trapezoid typically
somewhat concave centrally and enclosing several irregular parallel transverse
carinae; fore tibia with 16-26 spines arranged in irregular single to double rows
and a row of 6-8 spines at apex; outer apical lobe of hind tibia with 4-6 spines,
typically 5; hind coxae coarsely granular with weak oblique striation on the
inner dorsal surface; subdiscoideus not interstitial.

Abdomen: Petiole typically robust, dorsally granular to longitudinally stri-
gose apically and arched at base; tergite (2-3) always distinctly sculptured,
approximately the basal 5/11 to 1/2 strongly granular followed immediately
by a less strong or sometimes obscure granular to scaly-reticulate area on
about next 3/11 to transverse row of setae, the apical 3/11 beyond the setal
row always smooth; remaining tergites each with a well defined distinct trans-
verse punctulate band immediately in front of the row of setae, the apical por-
tion beyond the setal row smooth; punctulation extends onto the epipleurites but
is much weaker; tergite (2-3) weakly thickened laterally at base; ovipositor
slightly curved, 0.58-0.78x FWL.

Color: Body color basically dull to dark ferrugineous except that sternauli
and lower half of mesopleura are often much lighter, typically testaceous and
the basal half of tergite (2-3) and apical half of petiole are usually lighter,
especially in darker individuals; also, the venter and the apex of the gaster
tend to be lighter than the dorsum, often testaceous; eyes dark gray, with
rarely any evidence of iridescence in white light; antennae testaceous, becom-
ing darker apically; maxillary and labial palpi dark testaceous; coxae and tro-
chanters testaceous, never as dark as body, the hind coxae usually slightly
darker than the fore and mid coxae; femora and tibiae dark testaceous to ferru-
gineous, the tibiae distinctly lighter banded on basal 1/4 to 1/3; tarsi testaceous,

714 Contrib. Amer. Ent. Inst:, vol. 4, no..5, 1970

the ultimate tarsomere typically distinctly darker than others; fore wings
darkly infumated with a distinct lighter transverse band medially and apically
like floridanus (Fig. 3).

Males: Essentially agree with females in all important respects; number
of flagellomeres varies from 23-32 and the lateral margin of most gastral ter-
gites is thickened.

Remarks: According to the USNM Type Catalog (Type No. 2150), Ashmead
had a total of 27 individuals of 8. simillimus before him, including representa-
tives of both sexes. Comparing the numbers on the labels with those on Hop-
kins’ original manuscript for his 1892 paper, those determined as simillimus
include numbers 3095, 3207 and 3262, all reared from Agrilus bilineatus liv-
ing in white oak, and numbers 2557 and 2579 reared from an unknown host in
Abies nigra and a buprestid larva in an unstated host. However, in his descip-
tion, Ashmead does not note the latter hosts, stating simply, ‘‘Bred May 18
and 25, 1892, from Agrilus bilineatus living in White Oak stump.’’ Also, the
dates given agree only with numbers 3207 and 3262, respectively. Of four
pins bearing these numbers (with a total of 12 specimens), only one also has a
handwritten Ashmead label ‘‘Spathius simillimus Ashm.’’ (No. 3262).

This background is important in order that the name simillimus may be
properly and positively associated, since the 7 specimens on 4 pins labeled
with numbers 2579 and 2557 belong not to simillimus, but to the new species,
Spathius comes. I base my decision as to which should be called simillimus
primarily upon Ashmead’s statement of only the one host and the presence of
his determination label on but one pin as noted above, since his description,
with one exception, could apply equally well to either species. The exception
is his statement ‘‘ovipositor as long or very little shorter than the body. ’’

This would apply only to comes which has a distinctly longer ovipositor (OSL/
FWL 0.91-1.57) and not to simillimus as I define it (OSL/FWL 0. 58-0. 78).
However, on the basis of the above evidence I choose to ignore this discrepancy
and have selected as lectotype one of two females on one of the pins bearing
number 3207, and have designated it by a red ink mark at the base of the point.
All of the other specimens labeled with numbers 3207, 3095 and 3262 are thus
labelled as paralectotypes.

Aldrich (1921) gives a figure of Spathius simillimus. A specimen of this
species in the USNM bearing Hopkins No. 11254a has a label saying ‘‘type of
drawing’’ and may be assumed to be the specimen on which the drawing was
based. |
Diagnosis: The presence of distinct transverse punctulate bands on all gas-
tral tergites anterior to the setal row will usually distinguish this species from
most other North Americanforms. The very well developed sculpture on the
vertex and temples is also diagnostic, occurring in about equal development
only on S. floridanus which has the gaster nearly smooth. The form of the
sternaulus and mesopleuron is also distinctive, the carinae of the sternaulus
rarely distinctly extending onto the uniformly granular to scaly-reticulate
mesopleural disc or curving posteriorly tothe mesocoxal cavity as in flori-
danus, comes or pallidus which may sometimes be confused with it.

Distribution: Probably widespread in eastern North America north of
Virginia. However, as in the case of S. brunneus, I have seen only reared
material, consisting of five series in all, and representing Maine, Massachu-
setts, Virginia, West Virginia and New Brunswick, Canada. In addition,
Pennsylvania is listed (Muesebeck et al., 1951), Felt (1905) reports it from
New York, and Chittenden (1897) notes its abundance in the District of Colum-
bia. The reported occurrence of this species in Mississippi and Wisconsin

Matthews: Nearctic Spathius 715

(Muesebeck, 1967) is possibly erroneous, as I have not seen material to sup-
port it.

Hosts: Recorded hosts that I have been able to confirm are the buprestids
Agrilus bilineatus (Weber) and A. anxius Gory. The former was first recorded
by Hopkins (1892) from white oak and has been repeated by the following: Ash-
mead (1893), Hopkins (1893b, 1899), Chittenden (1897), Dalla Torre (1898),
Felt (1905), Viereck (1910), Muesebeck (1928, in Leonard), Doane et al. (1936),
and Muesebeck and Walkley (1951). Agrilus anxius in birch was first reported
as a host by Nash et al. (1951) and repeated in Muesebeck and Walkley (1951)
and Barter (1957). Balch and Prebble’s (1940) reference to Spathius sp. refers
to this species, as I have seen their specimens (CNC).

Hopkins (1892, 1899) records two additional species as hosts: Chryso-
bothris pusilla Cast. (Cerambycidae) in spruce and the weevil Cryptorhynchus
ferratus Say in sassafras. These are regarded as erroneous. The former is
no doubt a record for the new species, S. comes (Hopkins manuscript numbers
2557 and 2559). The latter Iam unable to substantiate or relate to any speci-
mens [I have seen.

Specimens of S. simillimus which I have seen bear Hopkins numbers 1717a,
11254 and 11847b. The first was reared from an unrecorded host in chestnut,
the second from Agrilus bilineatus in chestnut and the last from the cerambycid
Callidium aereum Newman in white oak, a new host record. Thus Spathius
simillimus seems comparatively specific in its host preferences, attacking
buprestid or cerambycid larvae in deciduous trees, especially birch and white
oak.

21. Spathius floridanus Ashmead
(Figs. 1-3, 24)

Spathius floridanus Ashmead, 1893, Can. Ent. 25: 71. [Type locality:
Florida: Jacksonville (holotype female in USNM, Type No. 52646)].

Females: Body length, 2.2-6.8 mm; FWL, 1.9-4.2 mm; ovipositor sheath
length, 1.3-4.6 mm. Other measurements and ratios in Table 6.

Head (Figs. 1 and 2): Flagellomeres 23-39, face coarsely transversely
strigose-granular; frons strongly transversely striate to rugose-granular,
with at least the granular sculpture always extending to eyes; vertex relatively
broad, OOD distinctly greater than OED (Fig. 24), and prominently and com-
pletely granular to transverse strigose-granular, extending undiminished to
eyes, frons, and onto entire temple nearly to malar space; ocellar triangle
weakly raised, granular.

Thorax: Lateral areas of prothorax variously defined, usually becoming
obscured posteriorly in the longitudinal granular strigosity of remainder of
propleura; pronotum dorsally broad with several fine slightly oblique trans-
verse striations usually visible anterior to where the lateral areas converge;
notauli deep, crenulate anteriorly meeting in a moderately depressed area of
strong rugosity extending to the posterior margin of the mesonotum; mesono-
tum abruptly declivous, the lobes uniformly granular; scutellar furrow relative-
ly broad with irregular variously developed cross carinae, these rarely simple
and distinct but with various transverse connections and rugosity at least par-
tially obscuring them; scutellum broad, moderately convex, the disc evenly
granular, becoming weakly rugose at posterior margin; mesopleural disc
variable, usually scaly-reticulate to nearly smooth anteriorly above sternau-

76 Contrib. Amer. Ent. Inst. , vol. 4, no. 5, 1970

lus, this giving way to irregular weak more or less longitudinal rugosities
immediately above posterior portion of sternaulus, with stronger longitudinal
granular strigosity occuring more dorsally beneath the wings; sternaulus vari-
ously developed, from a very weak granular longitudinal impression to occa-
sionally a sharply infolded steep-sided groove, but typically moderately im-
pressed and crossed by several poorly developed, irregular carinae some of
which may appear to extend weakly onto mesopleural disc or curve posteriorly
towards mesocoxal cavities; mesosternum scaly-reticulate adjacent to stern-
aulus, to nearly smooth; prepectal carina complete; prepectal area more or
less smooth with weak surface irregularity and sparse setae; propodeal cari-
nae distinct, and sharply defined; dorsal areas granular to rugose; basal cari-
na typically shorter than its forks, resulting ina relatively broad and elongate
areola which typically encloses several weaker reticulate cross carinae;
costulae well defined, angled somewhat anteriorly; area petiolaris small but
bounded by very thick and prominent carinae laterally; fore tibia with 13-25
spines arranged in an irregular single to double row and a row of 5-9 at the
apex; outer apical lobe of hind tibia with 2-4 spines, typically 3 situated some-
what anteriorly; subdiscoideus not interstitial.

Abdomen: Petiole moderately stout, arched at base, with granular irregu-
lar longitudinal strigosity dorsally; tergite (2-3) variable in degree of sculp-
ture developed, from virtually smooth and shining with perhaps very faint
traces of scaly-reticulate sculpture to distinct area of granular to scaly-
reticulate sculpture on basal 3/8 to 1/2, rarely with any sculpture present be-
yond the basal half; remaining tergites typically smooth but larger individuals
sometimes with weak indications of punctulate or faint scaly-reticulate trans-
verse bands anterior to the setal row on some tergites; epipleurites usually
very faintly punctulate; lateral margin of tergite (2-3) weakly thickened at base;
ovipositor varies from 0. 65-1. 11x FWL.

Color: Body color typically light ferrugineous, but considerable variation
exists from basically testaceous to dark ferrugineous, the gaster varying from
entirely shining honey yellow to entirely dark shining ferrugineous or with the
basal 1/2 to 2/3 of tergite (2-3) lighter than remainder of gaster or occasionally
the gaster with two to four subcuticular narrow transverse ferrugineous bands;
also the propodeum and petiole may be distinctly lighter than the gaster and re-
mainder of thorax, or the propodeum, petiole and part or all of the gaster may
be lighter than the remainder of the thorax; legs usually more or less concolor-
ous with body or slightly lighter, especially the trochanters and tarsi; tibiae
distinctly lighter banded on about basal 1/4-1/3; ultimate tarsomere usually at
least slightly darker than preceding tarsomeres; antennae testaceous to light
ferrugineous at base, gradually darkening apically; ovipositor sheaths testa-
ceous basally, becoming darker brown apically; eyes typically gray, rarely
with a faint iridescent hue in white light; fore wing (Fig. 3) distinctly banded
and darkly infumated, veins and stigma (except lighter basal 1/3) brown.

Males: Essentially parallel the variation in the females in sculpture and
color, although smaller males often have the intensity of the sculpture some-
what reduced on the head and mesopleura; however, the most notable differ-
ence is in the amount and intensity of sculpture on tergite (2-3), which in
males tends to regularly cover more than the basal 1/2, often extending to the
transverse setal row or over about 3/4 of the tergite; also gaster color beyond
the basal 2/3 of tergite (2-3) is frequently much darker, to nearly piceous.

Remarks: This robust species, based on a single female, was never for-
mally described beyond the characters given by Ashmead (1893) in his ‘‘Table
of Species’’, and has heretofore been of uncertain status. In part, this was

Matthews: Nearctic Spathius T7

due to its having been confused with S. trifasciatus Riley (see Remarks section
under that species) and in part because the holotype is a large specimen and in
some respects atypical for the species. In addition to the locality label, the
holotype has a label ‘‘Spathius floridanus Ashm.’’ in Ashmead’s handwriting
and the red ‘‘USNM Type No. 52646’’ label.

Diagnosis: This species is most closely related to Spathius simillimus,
from which it may be most readily distinguished by the absence of well devel-
oped transverse punctulate bands anterior to the row of setae on each of the
gastral tergites. In addition, the more rugose scutellar furrow, the usual
presence of more or less longitudinal irregularity on the mesopleural disc
immediately above the sternaulus, at least posteriorly (compared to the uni-
formly granular or scaly-reticulate mesopleural disc of simillimus) and the
presence of fewer spines on the outer apical lobe of the hind tibia (typically 3,
versus 5 in simillimus) will assist in distinguishing the two.

Distribution: Throughout eastern North America, from Florida to Ontario
and Quebec, Canada, and west to Texas, Kansas and Minnesota. The over
300 specimens which I have seen represent nearly every state in this region.

Biology: Spathius floridanus appears to differ from the majority of the
known Spathius species in that several individuals are usually reared from a
single host larva. I reared four individuals from a cluster of cocoons ina
cerambycid burrow under oak bark; a similar cluster of 9 cocoons is pinned with
Hopkins No. 1460c, and another cluster of 7 cocoons is pinned with Hopkins No.
9909b'. Comments on the cards in the Hopkins file often state ‘‘ex mass of
parasite cocoons in larval mines.’’ In each of the above three cases, the
reared adults include both sexes, suggesting that this is probably a case of
multiple oviposition.

Hosts: The only host previously recorded is the buprestid Chrysobothris
femorata (Oliv.) (Brimley, 1942; Muesebeck and Walkley, 1951). However,
in the U. S. National Museum are long series of Spathius floridanus from sev-
eral hardwoods, reared under numerous Hopkins numbers which often list hosts
and associated borers. These may be summarized as follows: from oak
(Quercus spp.) (Hopkins Nos. 1460c, 1421c, 3016c, 3730e, 11861k and 12858a),
hosts include Chrysobothris femorata and the cerambycid Xylotrechus colonus
F.; from hickory (Hicoria spp.) (Hopkins Nos. 1489f, 3447c, 3636e, 3651d and
10975c1), hosts include Chrysobothris sp. and the weevil Magdalis olyra Hbst. ,
with several additional associated borers listed under No. 3636e; from chestnut
(Castanea dentata)(Hopkins Nos. 10277b & c, 10701la-k, 11352al, and 11353),
hosts include C. femorata and X. colonus; for black walnut (Juglans nigra)
(Hopkins No. 3640a-j), hosts are not specified but several associated ceramby-
cids and buprestids are listed; for wild cherry (Prunus sp.) (Hopkins No.
10737i), parasitic on Chrysobothris sp.; for bald cypress (Taxodium distichum)
(Hopkins Nos. 1732c and 1640e), host is unknown. Several additional Hopkins
numbers are represented, but not listed here as they give no host tree or in-
sect data.

SPECIES INQUIRENDA
Spathius honestor Say, 1828, Contrib. Maclur. Lyc. Phila. 1: 78.

The holotype specimen of this species is no longer extant, and the original
description is inadequate to associate it, hence this species cannot be placed.
Ashmead (1893) apparently did not know it, as he omitted it from his key. No
hosts are recorded for this species.

718 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

SUMMARY

The braconid genus Spathius, ectoparasitic upon wood-boring beetle larvae,

is revised for North America. Keys to both sexes and complete descriptions
for the 21 recognized species are provided. Lectotypes are designated for the
following species: brachyurus Ashm., brunneus Ashm., claripennis Ashm.,
longipetiolatus Ashm., sequoiae Ashm, and simillimus Ashm. New sc meni
mies are: S. dignus Muesebeck and Walkley under S. brachyurus Ashm. ;
S. californicus Ashm. and S. brunneri Viereck under S. sequoiae Ashm. ; : and
S. claripennis Ashm. and S. tomici Ashm. under S. canadensis Ashm. Newly
described Spathius species include: aphenges, benefactor, brevipalpus, calli-
gaster, comes, elegans, evansi, impus, marshi, parvulus and stigmatus.

A revised list of all verified previous and new host records includes repre-
sentatives from six coleopterous families in both deciduous and coniferous
trees; analysis indicates that specificity tends to be more to host habitat than
to host taxon. Evidence for multiple oviposition in a single host is reported
for two species, S. brachyurus and 8. floridanus.

Evolutionary trends in the genus -are discussed, and aspects of the biology
reviewed, with particular reference to the Nearctic forms. Possible phyletic
relationships are considered and the status of the four described spathiine
fossils is clarified.

The long association of the name Spathius canadensis with parasites reared
from elm bark borers is shown to be erroneous; rather, the status of S. cana-
densis is redefined to be a species which typically attacks scolytids in various
coniferous trees. Most elm-reared specimens will belong to the new species,
S. benefactor.

Matthews: Nearctic Spathius 79

LITERATURE CITED*

Aldrich, J. M. 1921. The division of insects in the United States National
Museum. Smithson. Ann. Rep. for 1919, pp. 367-379.

Ashraf, M. and M. M. Berryman. 1969. Biology of Scolytus ventralis (Coleop-
tera: Scolytidae) attacking Abies grandis in northern Idaho. Melanderia
2: 1-23.

Ashmead, W. H. (1888) 1889. Descriptions of new Braconidae in the collection
of the U. S. National Museum. Proc. U. S. natn. Mus. 11: 611-671.

1891. Descriptions of some new Canadian Braconidae. Can. Ent.
23: 1-7.

1893. Descriptions of new braconids bred by Prof. A. D. Hopkins.
Can. Ent. 25: 67-79.

Ayyar, P. N. K. and P. S. Narayanaswami. 1940. On the biology of Spathius
vulnificus Wilk. , a possible effective parasite of Pempheres affinis in
South India. Indian J. Ent. 2: 79-86.

Balch, R. E. and J. S. Prebble. 1940. The bronze birch borer and its rela-
tion to the dying of birch in New Brunswick forests. For. Chron. 16:
179-201.

Barter, G. W. 1957. Studies of the bronze birch borer, Agrilus anxius Gory,
in New Brunswick. Can. Ent. 89: 12-36.

Beal, J. A. and C. L. Massey. 1945. Bark beetles and ambrosia beetles
(Coleoptera: Scolytidae): with special reference to species occurring
in North Carolina. Bull. Sch. For. Duke Univ. 10: 1-178.

Bedard, W. D. 1938. An annotated list of the insect fauna of Douglas fir
(Pseudotsuga mucronata Rafinesque) in the northern Rocky Mountain
region. Can. Ent. 70: 188-197.

Beyr, J. 1961. Ein Beitrag zur Kenntnis der morphologischen Variabilitat bei
Spathius exarator L. Verh. XI. Int. Kongr. Ent., Wien 1960 1: 229-232.

Blackman, M. W. and H. H. Stage. 1918. Notes on insects bred from the
bark and wood of American larch - Larix laricina (DuRoc.) Koch. Tech.
Publs N.Y. St.. Coll. For. 18: 9-115.

1924. On the succession of insects living in bark and wood of dying,
dead and decaying hickory. Tech. Publs N.Y. St. Coll. For. 24: 3-269.

Brimley, C. S. 1942. Supplement to Insects of North Carolina. N.C. Dep.
Agric., Raleigh, 39 pp.

Brues, C. T. 1910. The parasitic Hymenoptera of the Tertiary of Florissant,
Colorado. Bul. Mus. comp. Zool. Harv. 54: 1-125.

1927. Observations on wood-boring insects, their parasites and other
associated insects. Psyche 34: 73-90.

Bushing, R. W. 1965. A synoptic list of the parasites of Scolytidae (Coleoptera)
in North America north of Mexico. Can. Ent. 97: 449-492.

Bushing, R. W. and D. E. Bright. 1965. New records of hymenopterous para-
sites from California. Scolytidae (Coleoptera). Can. Ent. 97: 199-204.

Cameron, P. 1905. On the phytophagous and parasitic Hymenoptera collected
by Mr. E. Ernest Green in Ceylon. Spolia zeylan. 3: 67-97.

Chamberlin, W. J. 1939. The bark and timber beetles of North America north
of Mexico. Ore. St. Coll. Coop. Assn., Corvallis. 513 pp.

Chittenden, F. H. 1897. The two-lined chestnut borer (Agrilus bilineatus
Weber). Circ. Bur. Ent. U. S. Dep. Agric. (ser. 2) 24: 1-8.

*Abbreviations of periodicals follow World List of Scientific Periodicals, fourth
edition, 1963.

80 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

Clausen, C. P. 1940. Entomophagous Insects. McGraw-Hill, New York,
688 pp.

Craighead, F. C. 1950. Insect enemies of eastern forests. Misc. Publs U.S.
Dep. Agric. No. 657, 679 pp.

Dalla Torre, C. G. 1898. Catalogus Hymenopterorum. 4. Braconidae. G.
Engelmann, Leipzig, 323 pp.

De Leon, D. 1952. Insects associated with Sequoia sempervirens and Sequoia
gigantea in California. Pan-Pacif. Ent. 23: 75-91.

Doane, R. W., E. C. Van Dyke, W. J. Chamberlin and H. E. Burke. 1936.
Forest Insects. McGraw-Hill, New York, 463 pp.

Doutt, R. L. 1959. The biology of parasitic Hymenoptera. A. Rev. Ent. 4:
161-182.

Eady, R. D. 1968. Some illustrations of microsculpture in the Hymenoptera.
Proc. R. ent. Soc. Lond. (A) 43: 66-72.

Felt. E. P. 1906. Insects affecting park and woodland trees. Mem. N.Y. St.
Mus. 8(2): 333-887.

1915. 29th Report of the State Entomologist, 1913. Bull. N.Y. St.

Mus, 175: 1-257. |

Fischer, M. 1966. Studien tiber Braconiden (Hymenoptera). Z. angew. Zool.
53: 215-234, :

Fiske, W. F. 1907. Notes on the insect enemies of wood boring Coleoptera.
Proc. ent. Soc. Wash. 9: 23-27.

Foerster, A. 1862. Synopsis der Familien und Gattungen der Braconen. Verh.
naturh. Ver. preuss. Rheinl. 19: 225-288.

Gahan, A. B. andS. A. Rohwer. 1917, 1918. lLectotypes of the species of
Hymenoptera (except Apoidea) described by Abbe Provancher. Can. Ent.
49: 391-400; 50: 196-201.

Godwin, P. A. and T. M. Odell. 1965. Life history of the white pine cone
beetle, Conophthorus coniperda. Ann. ent. Soc. Am. 58: 213-219.

Heatwole, H. & D. M. Davis. 1965. Ecology of three sympatric species of
parasitic insects of the genus Megarhyssa (Hymenoptera: Ichneumonidae).
Ecology 46: 140-150.

Hetrick, L. A. 1940. Some factors in natural control of the southern pine
beetle, Dendroctonus frontalis Zimm. J. econ. Ent. 33: 554-556.
Hoffman, C. H. 1938. Notes on Leperisinus aculeatus (Say) and its parasites

(Coleoptera: Scolytidae). J. econ. Ent. 31: 118-119.
Hopkins, A. D. 1892. Some bred W. Virginia Braconidae. Insect Life 4:
256-259.
18938a. Catalog of West Virginia Scolytidae and their enemies with
list of trees and shrubs attacked. Bull. W. Va. agric. Exp. Stn. 31:
119-168.
1893b. Catalogue of West Virginia forest and shade tree insects.
Bull. W. Va. acric. Exp, Stn. 32: (71-251.
1899. Report on investigations to determine the cause of unhealthy
conditions of the spruce and pine from 1880-1893. Bull. W.Va. agric.
Exp. Stn. 56: 197-461.
Ishii, T. 1953. On the biology of Spathius fuscipennis Ashmead, a rice stem
borer parasite, in the Philippine Islands. Oyo Dobuts. Zasshi 18: 95-102.
Iwata, K. 1959. The comparative anatomy of the ovary in Hymenoptera.
Part Ill. Braconidae (including Aphidiidae) with descriptions of ovarian
eggs, Kontyu 27: 231-238.
Kaston, B. J. 1937. Notes on hymenopterous parasites of elm insects in
Connecticut. Bull. Conn. agric. Exp. Stn. 396: 351-361.

Matthews: Nearctic Spathius 81

1939. The native elm bark beetle Hylurgopinus rufipes (Eichhoff) in
Connecticut. Bull. Conn. agric. Exp. Stn. 420: 1-39.

Kaston, B. J. and W. B. Becker. 1936. Spathius canadensis Ashm., a para-
site of Hylurgopinus rufipes (Eichh.). J. econ. Ent. 29: 807.

Lyngnes, R. 1955. Zur Kenntnis der Biologie von Spathius exarator L. (Hym.,
Bracon.). Z. angew. Ent. 38: 73-81. ae

1960. Shape and function of the ovipositor in the three Hymenopter-
ous species: Ephialtes extensor Thoms. (Ichneumonidae), Spathius
exarator L. (Braconidae), and Plutothrix coelius Walk. (Chalcidae).
Norsk ent. Tidsskr. 11: 122-134.

MacAloney, H. J. 1930. The white pine weevil (Pissodes strobi Peck) - its
biology and control. Bull. N.Y. St. Coll. For. 3: 1-87.

Marsh, P. M. 1963. A key to Nearctic subfamilies of the family Braconidae
(Hymenoptera). Ann. ent. Soc. Am. 56: 522-527.

1965. The Nearctic Doryctinae. I. A review of the subfamily with
a taxonomic revision of the tribe Hecabolini (Hymenoptera: Braconidae).
Ann. ent. Soc. Am. 58: 668-699.

Matthews, R. W. and P. M. Marsh. 1969. On the status of Cresson’s
Stenophasmus species from Cuba (Hymenoptera: Braconidae: Doryctinae).
Proc. ent. Soc. Wash. 71: 94-98.

Muesebeck, C. F. W. 1928. Family Braconidae. pp. 895-920. In: Leonard,
M. D., List of the insects of New York. Cornell Univ. agric. Exp. Stn.
Mem. No. 101, 1100 pp.

1967. Family Braconidae. pp. 27-60. In: Krombein, K. V. and
B. D. Burks, Hymenoptera of America North of Mexico. Synoptic
Catalog. Agriculture Monogr. No. 2, suppl. 2, 584 pp.

Muesebeck, C. F. W. and L. M. Walkley. 1951. Family Braconidae.
pp. 90-184. In: Muesebeck, C. F. W., K. V. Krombein and H. K.
Townes, Hymenoptera of America North of Mexico. Synoptic Catalogue.
Agriculture Monogr. No. 2, 1420 pp.

Nash, R. W., E. J. Duda and N. H. Gray. 1951. Studies on extensive dying,
regeneration, and management of birch. Bull. Me. Forest Serv. 1.

Nees von Esenbeck, C. G. 1818. Appendix ad J. L. C. Gravenhorst conspectum
generum et familiarum Ichneumonidum, genera et familiam Ichneumoni-
dum adscitorum exhibens. Nova Acta Acad. Caesar. Leop. Carol. 9:
299-310.

Nixon, G. E. J. 1943. A revision of the Spathiinae of the Old World (Hy menop-
tera, Braconidae). Trans. R. ent. Soc. Lond. 93: 173-456.

Pechuman, L. L. 1937. An annotated list of insects found in the bark and
wood of Ulmus americana L. in New York State. Bull. Brooklyn ent.
Soc, 32: 8-21.

Pierce, W. D. 1908. A list of parasites known to attack American Rhyncophora.
J. econ. Ent. 1: 380-396.

Provancher, L. 1880. Faune Canadienne: Les Insectes Hymenopteres.
Naturaliste can. 12: 161-180.

1886. Additions a la Faune Canadienne - Hymenopterologique.
Naturaliste can. 16: 121-157.

1888. Additions a la Faune Hymenopterologique. Supplement aux
additions aux Hymenopteres. Naturaliste can. 18: 370-397.

Riley, C. V. 1873. Fifth annual report on the noxious, beneficial and other
insects of the State of Missouri. Regan & Carter, Jefferson City, Mo.

160 pp.

82 Contrib. Amer. Ent. Inst., vol. 4, no. 5, 1970

Riley, C. V. and L. O. Howard. 1890. Some of the bred parasitic Hymenop-
tera in the National Collection. Insect Life 2: 348-353.

Robert, A. 1951a. Notes biologiques sur les principaux insectes parasites des
charancons de l’orme: Eubadizon magdali Cress., Spathius canadensis
Ash. (Braconides), Trigonura tarsata D. T. , Trigonura sp. (Chalcidides,
Hymenopteres). Annls ACFAS 17: 138-142.

1951b. Role respectif de chacune des especes parasites dans la
reduction des charancons de Porme. Annls ACFAS 17: 143-147.

(1960) 1961. Les hymenopteres parasites des charancgons de l’orme
et leur influence sur leurs hotes dans le Quebec. Ann. ent. Soc. Queb.
6: 122-143.

Russo, C. V. 1938. Contributo alla conoscenza dei Coleotteri Scolitidi
Fleotribo: Phloeotribus scarabaeoides (Bern.) Fauv. Parte seconda.
Biografia, simbionti, dannie lotta. Boll. Lab. Entomol. Agr. Portici
2: 1-420.

Say, T. 1828. Descriptions of new species of By emo ane of the U. S.
Maclurian Lyceum, Phila. 1: 67-83.

Schedl, K. E. 1932. Parasites reared from forest insects in 1929. Can. Ent.
64: 1-2.

Scudder, S. H. 1890. The tertiary insects of North America. Rep. U. S.
geol. Surv. 13: 1-734.

Shenefelt, R. D. 1968. Braconidae described by 1? Abbe Provancher.
Naturaliste can. 95: 627-656.

Smith, F. 1859. Catalogue of hymenopterous insects collected by Mr. A. R.
Wallace at the islands of Aru and Key. J. Linn. Soc. 3: 132-178.

Snodgrass, R. E. 1941. The male genitalia of Hymenoptera. Smithson.
misc. Collns 99 (14), 86 pp.

Statz, G. 1936. Ueber alte und neue fossile Hymenopterenfunde aus den
tertiiren Ablagerungen von Rott am Siebengebirge. Decheniana 93:
256-312.

1938. Neue Funde parasitischer Hymenopteren aus dem Tertiar von
Rott am Siebengebirge. Decheniana 98A: 71-154.

Tobias, V. I. 1967. A review of the classification, phylogeny and evolution of
the family Braconidae (Hymenoptera). Ent. Obozr. 46: 645- 669.
[translation in Ent. Rev., Wash. 46: 387-399. |

Townes, H. 1962. Host selection patterns in some Nearctic ichneumonids.
Verh. XI. Int. Kongr. Ent., Wien 1960 2: 738-741.

Viereck, H. L. 1910. Hymenoptera. pp. 579-700. In: Smith, J. B., The
insects of New Jersey. A. Rep. New Jers. St. Mus. (1909), 888 pp.

1912. Contributions to our knowledge of bees and ichneumon-flies,
including description of twenty-one new genera and fifty-seven new spe-
cies of ichneumon-flies. Proc. U. S. natn. Mus. 42: 613-648.

Walker, F. 1874. Descriptions of some Japanese Hymenoptera. Cist. Ent.

1; 301-310.

Matthews: Nearctic Spathius 83

Figure 5. Spathius marshi, sp. n., lateral view of 2; arrow above points to projecting
carina separating areola and area petiolaris; other arrow points to sternaulus.

Figure 6. Portion of antenna of Spathius stigmatus, sp. n., showing basal flagellomeres,
pedicel and apical part of scape.

Figure 7 and 8. Heads of Spathius brevipalpus, sp. n. andS. brachyurus Ashmead,
respectively, lateral view, drawn at same magnification.

Figures 9-13. Wings of Spathius, all drawn at same magnification: 9. S. marshi,
sp. n., 2; 10. S. brachyurus Ashmead, ?; 11. 8S. brevipalpus, sp. n., &;
12. S. stigmatus, sp. n., 9; 13. S. stigmatus, sp. n., % hind wing.

fo)

15

Figure 14. Thorax of Spathius brunneus Ashmead, lateral view; arrow points to longi-
tudinal carina below sternaulus.

Figures 15 and 16. Hind legs of males of Spathius trifasciatus Riley and S. sequoiae
Ashmead, respectively.

Figure 17. Head and thorax of Spathius calligaster, sp. n., lateral view, 2; arrow above
points to transverse swelling of prothorax, other arrow to incomplete prepectal
carina.

Figure 18. Outer apical lobe of hind tibia of Spathius impus, sp. n.

Figure 19. Fore wing of Spathius calligaster, sp. n., 9°.

Figure 20. Male genitalia of Spathius longipetiolatus Ashmead.

Figure 21. Fore wing of Spathius evansi, sp. n.,-°.

Figure 22. Spathius elegans, sp. n., lateral view of ¢.

Figure 23. Outer apical lobe of hind tibia of Spathius laflammei Provancher.

Matthews: Nearctic Spathius 85

Figures 24 and 25. Heads of Spathius floridanus Ashmead and S. parvulus, sp. n.,
respectively; dorsal views at same magnification.

Figures 26 - 31. Fore wings of female Spathius, all drawn at same magnification:
26. S. aphenges, sp. n.; 27. S. sequoiae Ashmead; 28. S. pallidus Ashmead;
29. S. trifasciatus Riley; 30. S. parvulus, sp. n.; 31. 5S. comes, Sp. n.

86

INDEX TO SPATHIINES

An asterisk indicates a synonym, misspelling, or nomen nudum.

*abdominalis 29 |

aphenges 12, 15, 17, 18, 19, 22, 25,
45, 47, 51-53

aristaeus 5

benefactor 12, 15, 16, 17, 18, 19,
23, 26, 49,50, 61-64, 72

brachyurus 12, 14, 16, 17, 18, 19,
20, 23, 26-29, 34, 50

*brevicaudus 26

brevipalpus 5, 12, 14, 18, 19, 20,
28, 29-30, 34

*bruneus 70

*brunneri 44, 46, 47, 71

brunneus 12, 15, 17, 19, 21, 24, 66,
67, 68, 70-72, 74

*californicus 44, 46, 47

calligaster 7, 9, 12, 14, 17, 18, 21,
24, 35-38, 65

canadensis 12, 15, 16, 18, 19, 22,
25, 26, 28, 45, 46, 47-51, 55, 59,
63

cassidoris 32

*claripennis 47, 49, 50, 63

*clavipennis 48

comes 12, 15, 17, 22, 25, 57, 59-61,
69, 74, 75

critolaus 19

*dignus 26, 27, 28

elegans 12, 14, 17, 21, 24, 36,
39-42

erigone 10

*Fuspathius 10

*Eustalocerus 26

evansi 12, 14, 17, 21, 24, 36, 42-43

exarator 8, 10, 12, 16

*fasciatus 26, 27

floridanus 12, 16, 17, 18, 22, 25, 29,
99, 65, 68, 72, 74, 715-77 |

fuscipennis 16

honestor 77

impus 12, 15, 18, 23, 26, 50, 51, 54,
06-59, 61

laflammei 12, 15, 16, 17, 18, 22, 26,
34, 63; 64-65, 72

longicornis 11

longipetiolatus 12, 14, 17, 21, 24, 36,
38-39

macroradialis 11

marshi 12, 14, 17, 20, 24, 30-32, 34

pallidus 12, 15, 16, 18, 23, 26, 50,
03-99, 57, 59, 74

parvulus 12, 15, 18, 23, 25, 26, 52,
54, 55-56, 57; 59

pedicularis 11

petrinus (Ichneumon) 11, 12

*pollidus 53

*Rhacospathius 10

*Rhopalophorus 26, 27

rubidus 16

ruficeps 10

sequoiae 12, 15, 17, 18, 19, 23, 25,
44-47, 49, 50, 52

simillimus 12, 15, 17, 22, 25, 42, 50
O09, 61, 65, 72-75, 77

*Stenophasmus 10

stigmatus 8, 12, 14, 16, 17, 18, 19,
20, 23, 32-35, 65, 68, 69

striolatus 10

*tomici 48, 49, 51, 59, 63

trHasemius 9, 12, 15, 16, 19, 21, 24,
35, 46, 66-69, 71, 77

*unifasciatus 9, 46, 66, 68, 69

vulnificus 12, 17

’

LE TEE ESE STE Oa SS SS TSS SS SE SST AEGIS

Publication dates for numbers of volumes 1-3 of Contributions of American

Entomological Institute.

Volos Vol. 2 Vol. 3

no. 1, October 7, 1964 no. 1, August 7, 1967 no. 1, April 30, 1968
no. 2, June 21, 1965 no. 2, July 25, 1967 no. 2, November 8, 1968
no. 2a, June 6, 1967 no. 3, July 25, 1967 no. 3, August 27, 1968
no. 3, September 15, 1965 no. 4, April 30, 1968 no. 4, November 8, 1968
no. 4, June 25, 1965 no. 5, April 30, 1968

no. 5, September 9, 1965

no. 6, August 12, 1966

no. 7, March 28, 1967

no. 8, June 5, 1967

Se:
Vihspina

ene

ahs

15:

%
fan

Nosits

(Continued from inside front cover)
Parts of Volume 4, with prices

Mosquito studies (Diptera, Culicidae), 1968.

X. Peters, T. Michael. Dixinae originally described from North America. 7 pages.
XI. Belkin, John N., Robert X. Schick, and Sandra J. Heinemann. Mosquitoes
originally described from Argentina, Bolivia, Chile, Paraguay, Peru, and Uruguay.
29 pages. Price: $1.25, postpaid.

Mosquito studies (Diptera, Culicidae), 1969.

XII. Berlin, O. G. W. A revision of the Neotropical subgenus Howardina of Aedes.
190 pages, 58 figures. Price: $6.00, postpaid.

Carlson, Robert W. and Fred B. Knight. Biology, taxonomy, and evolution of four
sympatric species of Agrilus beetles (Coleoptera: Buprestidae). 105 pages, 47 figures.
Price: $3.50, postpaid.

Mosquito studies (Diptera, Culicidae), 1969.

XIII. Barr, Ralph A. and Sylvia Barr. Pupae of the genus Psorophora in America north
of Mexico. 19 pages, 11 figures.

XIV. Berlin, O. G. W. Redescription of Toxorhynchites (T.) gigantulus from the
Philippines. 7 pages, 2 figures. ae

XV. Zavortink, Thomas J. A new species of treehole breeding Anopheles from the south-
western United States. 12 pages, 4 figures.

Price: $1.25, postpaid.

Matthews, Robert W., 1970. A revision of the genus Spathius in America north of Mexico
(Hymenoptera, Braconidae). 86 pages, 31 figures. Price: $3.50, postpaid.

i

a

cs

%
oe

tie ded Ned

Sa
tea

Ww, .weys mx yf 5 uy) =, “CN
Cae = re . = Xe ary oa” mi, Sy. 2
; wane oe Noss ‘ia oO INOSRY
TuUVysS Li! BRARI ES SMITHSONIAN _ NOILNALILSNI NVINOSHLIWNS S31uVe
Jt = a = ‘ee ‘ i a a
uw eee “wo Oo 12; KUSVAP
a S 4 Ze > ae os
20 - Gy, Pa Bes
as Ly ines h -
m hs “if die = =
w Re ee = =
NOILNLILSNI NVINOSHLINS INST
oa oo = z a .
et a at Be aa” ae Mae oO S
in 2 a in 3 a 7) |
© he peek oe pe rs NE
IYUVYUEIT LIBRARI ES SMITHSONIAN pNOlALIisnl NVINOSHLINS S3IYVY
See 6 * ears x eee . i 2
c <3 = - 2 SO <
Oo = Ce os ps rs lous .
—— AS ee oe ee —— ~ ee i
STITUTION NVINOSHLINS S3!IYVudiy INSTITU:
: Zz * = , i = ;
; Oo ee eas : ~~ <
— oe —<— ic . -
+ , — ve = i
he > ae > we
i= a a 9 = 2
iYvuaia LIBRARIES SMITHSONIAN INSTITUTION NVINOSHLINS S31uVvYy
205 = os n ne | “” = ” |
= a ee = = = iy
(2) SN ae oO x Baie mi ime.
E Nez eS AO, = Z
STITUTION NOILNLILSNI NVINOSHLIWS SSIYVYGIT LIBRARIES SMITHSONIAN INSTITU
ef ee = uJ - Ww e <5 wa
os od am = = tp, P Kam
o - a & te Yue
ue Ales ASV . | ~t mae + \OSd!
. Se fs = o “hon
HtYveaIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS S31uVe
ae 3 | =
— pal hy os Us os 2 Ka
8} fons i Ly, q - "it “ ris : Y
m o ‘if ee ph m - ens
be NOILNLILSNI NVINOSHLINS SSIYVYUEIT LIBRARIES SMITHSONIAN INSTITU!
ie 3 <a eee z we YW? ad ae
= — 3 = «St ‘ NS = < x
= Zz 4 eS SS <4 >
ZS gy a a 2
S E a. i Z
> Y = > e = x > >
1;uVUsgIiy LIBRARI ES SMITHSONIAN INSTITUTION NOILNIILSNI NVINOSHLIWS
“4 oc. a rd a Ate
—~ <<? a < ~ ae
c a eo ac a oc y
mikk © a aa) we cm When
O oo “. © | aur CC pee L150
= —t eae od Coke Pa '
NVINOSHLINS SJIYVYGIT LIBRARIES SMITHSONIAN INSTITU
> r = 8 z A J
= Ww . w ~ m8, oe SO
ri > > cS i> (F

ie gh
pe i

= ~ SENS = .
a > ae ~~,
, a
} ~ § ie Ne ‘ “
i ; 3 ‘ A :
j ogy *
ad :
= ¢ . es
—_ ; & |
j™ ‘ —"
» ? S ote 4 ee ‘Se fe 2 ea ;
4 Ne y Saacaait
J va
casey, Ie KF fp ’
—_ 3 ~
saad fe . 7 + 3)
4 y 'f P
» ~wehrte vy , . . : .
t¥ i bit pw Vibi‘s
j
/ - 1
_ ~ 4
z ; ‘5 % ‘ee ee . “ Bie
me § , i Las # - wl na ;
—_ . .
* se j ~
} ‘ry i
' ’ ; ; . F . LJ? ht
i H Boe i SS ‘ st 5 {
. —, -
ff al 4}, sé

3

i¥ } d a he & ~ . . +
i 3 . ‘

aig peu

apa riot

Be elspa
Reese

Seote) se

ala?)

Bay 245 ky

Wath ik agg aie

naianty

$NA AY

Cds pail eee ig aes

2 ENE ART

PA MAG pone aS by Saas call, Por i Y K ‘

Goudie

EN

be

Pe iq sips ap ‘
BA LANs SaiV IBN peneaee sees
BSA, CE SPN

aie ap ety si sprienel a:
Ay es ! Ps ee Ae pap Sp
pes (Eavancarrve AMES Y Re eases ae EY ge oo .

PAE ME AWS Ups hi

4 : r - TA ba AKT pel 4 : woe hse
AE NEA ala Pie A pe RMI 3 K v Wan

re 29 hia pi fe

one
bed em wre Geisha ray aR Ree 2 Bisips amen ok AREY

RY ee 2 nied it SU BRIsA

9 ib he ap esies BP i ape

iy ree
, ApS Tab ai Sp) shlWhip patie ogy Shah
i x iS 3 qi MALE Fp Gh set 5 ey at pe Sys eae ceili Pe Poh gle yaa
nh thts ; NMP APA std ty cies PE RAG j ek tel f E ’ v4 ; é : rite
| : Pe SPW ESE 1? Se Sci SEO SEH RBNA
RI Sy a ih ape. gs : lose ene ee
Oat det peck
EPR Fae Bay
ea os SANS, ae set: <iteits “Ba sigh a] 4 PSPS ry) sits SUSUR peed
ssa j i q j aise é i ? Hea B88 be Bena eeen POU TEE ABIES Sade
i 4 : Reape ie te ty ; ; pals a aces , ; : se : is : RGAE AM beasts 2h Ses ala Carey ee nes ee
Mey ik Aesth ay SAN ah “i ‘ J NER AF 3M My a dlp dab seaeets ah : ; ; : Meipist dials :
NGL AL GA, gee AM. fff) i re 4 NM SAO A F He, ‘f 5 “isin : 2 i Be ‘ Pee, be: ‘
A Abi £334 ay i Akay : B se i : i i ¥

eH =

MPa